HK1218582B - Method and device for detecting specific identification image in predetermined area - Google Patents

Description

A method and device for detecting a specific identification image in a predetermined area

Technical Field

The present application relates to the field of computer applications, and in particular to a method and device for obtaining a specific identifier in a standard card.

Background Art

The recognition of specific identifiers falls under the category of object detection. Specific identifiers are generally used to indicate certain characteristics of an object. For observers, identifiers can serve as a reference for identifying the object. Specific identifiers typically have a distinct appearance and a prominent spatial location, making them easier to identify. For bank cards, identifiers generally include the issuing organization's graphic logo and text logo. Recognizing these identifiers has the following important implications:

(1) The identification of identifiers can serve as an important reference for identifying the authenticity of bank cards.

Markers generally have clear size specifications and relative position markings. Legitimate bank cards will have markers printed in accordance with the standards. However, counterfeit bank cards are generally not strictly controlled in the printing of markers due to their low workmanship and low cost. By identifying the markers, the size ratio and relative position of the markers can be obtained, and further compared with the real standard bank cards, it can provide an effective reference for identifying the authenticity of bank cards.

(2) The identification of the identifier can provide a reference identifier for the detection and correction of bank cards.

Compared to the entire bank card, the marker has a uniform design and therefore generally has a clear edge marker. By identifying the marker's edge, we can derive the affine transformation of the current image based on this edge. This allows for image correction to facilitate subsequent accurate bank card detection and location. The term "border marker" here is not entirely accurate; the more accurate term is "edge marker."

(3) Identification of identifiers allows service providers to provide precise advertising push services based on specific identifiers.

Identifiers often indicate the specific identification of bank cards, such as the issuing organization and scope of use. Therefore, the identification of identifiers allows merchants to accurately push advertisements for the bank cards and improve the quality of service. In addition, after obtaining the identifier, combined with technologies such as augmented reality, multimedia advertising display can be achieved.

Object detection refers to finding the location of a specific identifier or object of interest in a given image or video. Depending on the specific usage scenario, the object detection system usually outputs the object's circumscribed rectangular box or precise outline.

Object detection can be broadly categorized into two methods: sliding window and generalized Hough transform. The sliding window method, currently the mainstream object detection method, uses a trained template to perform sliding scans across multiple scales and perspectives of the input image, then locates the window surrounding the target object by determining the maximum corresponding position.

The sliding window method used in existing object detection methods requires a large amount of computation and has relatively high computational complexity when analyzing and scanning data, resulting in a long detection time. This method is suitable for detecting objects with complex structures. However, it is clearly not suitable for identifying signs with relatively fixed structures.

Therefore, a technical problem that urgently needs to be solved by those skilled in the art is how to provide a method that can quickly and accurately identify a specific mark on a standard card.

Summary of the Invention

The present application provides a method for detecting a specific identification image in a predetermined area, so as to solve the problems of high computational complexity and large amount of calculation in the existing detection process.

The present application provides a method for detecting a specific identification image in a predetermined area, comprising:

Acquire an image of a predetermined area as an image of a target to be inspected;

Binarizing the target image to be inspected to obtain a binary image and an inverse image of the binary image, both of which are referred to as the target binary image to be inspected;

Calibrate the connected domain of the binary image of the target to be detected, and take the area where the number of pixels in the calibrated connected domain meets the set threshold range as the candidate connected domain area;

The image features of the candidate connected domain regions are obtained one by one, and the image features of the candidate connected domain regions are compared with the corresponding image features of the standard specific identification image, and the candidate connected domain regions with the same or closest comparison results are regarded as the locations of the specific identification images.

Preferably, acquiring the predetermined area or an image containing the predetermined area as the target image to be inspected includes:

Acquiring a preview image of the predetermined area or an image containing the predetermined area;

Calculating the area range of the preview image;

The preview image is framed within the region, the predetermined region or an image containing the predetermined region is acquired, and the image within the framed region is used as a target image to be inspected.

Preferably, calculating the area range of the preview image refers to obtaining virtual frame parameters for framing the preview image based on actual size parameters of the predetermined area and resolution parameters of the preview image, specifically:

Determining, based on the actual size of the predetermined area or the image containing the predetermined area, whether a value obtained by multiplying the height of the preview image by the actual width of the predetermined area or the image containing the predetermined area is greater than a value obtained by multiplying the width of the preview image by the actual height of the predetermined area or the image containing the predetermined area;

If so, the width of the virtual border is:

ScaledWidth＝CardWidthGdt×PreviewHeight/CardHeightGdt×DisplayRatio;

Wherein, CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; PreviewHeight is the height of the preview image; CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image;

The height of the virtual border is:

ScaledHeight＝PreviewHeight×DisplayRatio;

Where PreviewHeight is the height of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image;

If not, the width of the virtual border is:

ScaledWidth＝PreviewWidth×DisplayRatio;

Where PreviewWidth is the width of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image;

The height of the virtual border is:

ScaledHeight＝CardHeightGdt×PreviewWidth/CardWidthGdt×DisplayRatio;

Wherein, CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; PreviewWidth is the width of the preview image; CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image;

The upper boundary of the virtual border is half of the difference between the display screen height and the virtual border height; the lower boundary of the virtual border is half of the sum of the display screen height and the virtual border height; the left boundary of the virtual border is half of the difference between the display screen width and the virtual border width; and the right boundary of the virtual border is half of the sum of the display screen width and the virtual border width.

Preferably, the obtaining of the predetermined area or the image containing the predetermined area as the target image to be inspected includes: based on the specific position information of the specific identification image in the predetermined area or the image of the predetermined area, intercepting the area range containing the specific identification image as the target image to be inspected.

Preferably, the specific position information is the position of a pre-set specific identification image located in a predetermined area or on an image including a predetermined area.

Preferably, the specific position is the upper right corner or lower right corner or the upper left corner or lower left corner of the predetermined area or the image including the predetermined area.

Preferably, based on the specific position information of the specific identification image in the predetermined area or the predetermined area image, intercepting the area range containing the specific identification image as the target image to be inspected includes:

Establishing a predetermined area or a standard image template containing a predetermined area;

Traversing the standard image template according to the predetermined area or the image containing the predetermined area, and obtaining position information of the specific identification image in the predetermined area or the image containing the predetermined area through a correspondence between the standard image template and the predetermined area or the image containing the predetermined area;

The predetermined area or the image area containing the specific mark in the image containing the predetermined area is intercepted according to the position information as the target image to be inspected.

Preferably, the upper limit value of the threshold interval is the product of the width, height and upper limit coefficient of the acquired predetermined area image or the image containing the predetermined area; the lower limit value is the product of the width, height and lower limit coefficient of the acquired predetermined area image or the image containing the predetermined area.

Preferably, the comparing the image features of the candidate connected domain area with the corresponding image features of the specific identification image includes: using a single image feature comparison or a combined image feature comparison; wherein the image features include the volume ratio of the connected domain, the major axis direction of the ellipse circumscribed by the connected domain, the outer contour shape of the connected domain and/or the aspect ratio of the rectangle circumscribed by the connected domain.

Preferably, the image features of the candidate connected domain regions are obtained one by one, and the image features of the candidate connected domain regions are compared with the corresponding image features of the specific identification image, and the candidate connected domain regions with the same or closest comparison results are regarded as the regions where the specific identification image is located, including:

The candidate connected domain regions are screened using the volume ratio of the connected domain and the major axis direction of the connected domain circumscribed ellipse as combined image features, specifically:

Get the bounding box of the candidate connected region;

Calculating a volume ratio of pixels of the foreground color in the candidate connected domain within the bounding box to the area of the bounding box;

Determine whether the volume ratio is greater than a preset threshold range of a specific identification image; if not, discard the candidate connected domain area; if so, fill the candidate connected domain area;

Calculating the direction of the major axis of the ellipse circumscribed in the filled candidate connected region, and obtaining the angle between the major axis and the set reference axis;

It is determined whether the angle is within a predetermined threshold range. If so, the candidate connected domain area is regarded as the area where the specific identification image is located; if not, the candidate connected domain area is discarded.

Preferably, the method includes: after determining that the angle between the direction and the reference axis is within a predetermined threshold range, screening the candidate connected domain regions again based on the outer contour shape of the connected domain as a separate condition, specifically:

Filling the candidate connected component area to obtain a filled image;

Detecting the edge of the filled image to obtain an edge contour image of the filled image;

Respectively obtaining sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image, and selecting the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range;

Fitting the selected sampling pixel points to obtain line segment combination information of the edge contour image;

Determine whether the line segment combination information and intersection position information are the same as the line segment combination information and intersection position information of the standard specific identification image; if so, regard the candidate connected domain area as the specific identification image area; if not, output the undetected specific identification image area.

Preferably, obtaining sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image includes:

Scanning is performed on the condition that there are two intersection points with the filling image, and the intersection points are used as sampling pixel points;

The selecting of the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range includes:

Determine whether the distances between the upper and lower edges and the distances between the left and right edges of the obtained sampling pixel points are within a set threshold range. If so, select them; if not, discard them.

Preferably, the threshold range is maxDist±maxDist×5%.

Preferably, it includes:

Calculating an affine transformation matrix based on the position information of a specific marker in a predetermined area or an image containing the predetermined area and the position information of each intersection point;

Relevant information within the predetermined area is acquired according to the affine transformation matrix.

Preferably, the specific identification image is a logo graphic of a bank card association or a bank card issuing organization.

The present application further provides a device for detecting a specific identification image in a predetermined area, comprising:

an acquisition unit, configured to acquire the predetermined area or an image containing the predetermined area as an image of a target to be inspected;

A binarization processing unit, configured to perform binarization processing on the target image to be detected to obtain a binarized image and an inverse image of the binarized image, both of which are referred to as the target binary image to be detected;

A connected domain calibration unit is used to calibrate the connected domain of the binary image of the target to be detected, and to select the connected domain whose number of pixels in the calibrated connected domain meets the set threshold range as a candidate connected domain;

The image feature comparison unit is used to obtain the image features of the candidate connected domain one by one, and compare the image features of the candidate connected domain with the corresponding image features of the standard specific identification image, and regard the candidate connected domain with the same or closest comparison results as the location of the specific identification image.

Preferably, the acquiring unit includes:

a preview image acquiring unit, configured to acquire a preview image of the predetermined area or an image containing the predetermined area;

A calculation unit, configured to calculate the area range of the preview image;

The framing unit is used to frame the preview image within the area, obtain the predetermined area or an image containing the predetermined area, and use the image within the framed area as the target image to be inspected.

Preferably, the calculation unit is configured to obtain virtual frame parameters for framing the preview image according to the actual size parameters of the predetermined area and the resolution parameters of the preview image, specifically:

Determining, based on the actual size of the predetermined area or the image containing the predetermined area, whether a value obtained by multiplying the height of the preview image by the actual width of the predetermined area or the image containing the predetermined area is greater than a value obtained by multiplying the width of the preview image by the actual height of the predetermined area or the image containing the predetermined area;

If so, the width of the virtual border is:

ScaledWidth＝CardWidthGdt×PreviewHeight/CardHeightGdt×DisplayRatio;

Wherein, CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; PreviewHeight is the height of the preview image; CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image;

The height of the virtual border is:

ScaledHeight＝PreviewHeight×DisplayRatio;

Where PreviewHeight is the height of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image;

If not, the width of the virtual border is:

ScaledWidth＝PreviewWidth×DisplayRatio;

Where PreviewWidth is the width of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image;

The height of the virtual border is:

ScaledHeight＝CardHeightGdt×PreviewWidth/CardWidthGdt×DisplayRatio;

Wherein, CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; PreviewWidth is the width of the preview image; CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image;

The upper boundary of the virtual border is half of the difference between the display screen height and the virtual border height; the lower boundary of the virtual border is half of the sum of the display screen height and the virtual border height; the left boundary of the virtual border is half of the difference between the display screen width and the virtual border width; and the right boundary of the virtual border is half of the sum of the display screen width and the virtual border width.

Preferably, the acquiring unit includes:

The interception unit is used to intercept the area range containing the specific identification image as the target image to be inspected according to the specific position information of the specific identification image in the predetermined area or the predetermined area image.

Preferably, the specific position information is the position of a pre-set specific identification image located in a predetermined area or on an image including a predetermined area.

Preferably, the specific position is the upper right corner or lower right corner or the upper left corner or lower left corner of the predetermined area or the image including the predetermined area.

Preferably, the interception unit comprises:

An establishing unit, configured to establish a predetermined area or a standard image template containing the predetermined area;

a traversal unit, configured to traverse the standard image template according to the predetermined area or the image containing the predetermined area, and obtain position information of the specific identification image in the predetermined area or the image containing the predetermined area through a correspondence between the standard image template and the predetermined area or the image containing the predetermined area;

The target image to be inspected determining unit is configured to intercept the predetermined area or an image area containing a specific mark in the image containing the predetermined area according to the position information as the target image to be inspected.

Preferably, the upper limit value of the threshold interval is the product of the width, height and upper limit coefficient of the predetermined area or the image containing the predetermined area; the lower limit value is the product of the width, height and lower limit coefficient of the predetermined area or the image containing the predetermined area.

Preferably, the comparing the image features of the candidate connected domain area with the corresponding image features of the specific identification image includes: using a single image feature comparison or a combined image feature comparison; wherein the image features include the volume ratio of the connected domain, the major axis direction of the ellipse circumscribed by the connected domain, the outer contour shape of the connected domain and/or the aspect ratio of the rectangle circumscribed by the connected domain.

Preferably, it includes: a combined image feature screening unit, specifically:

An outer bounding box obtaining unit, used to obtain the outer bounding box of the candidate connected region;

a volume ratio calculation unit, configured to calculate a volume ratio of pixels of the foreground color in the candidate connected domain within the circumscribed frame to the area of the circumscribed frame;

A volume ratio threshold judgment unit is used to judge whether the volume ratio is greater than a preset threshold range of a specific identification image; if not, the candidate connected domain area is discarded; if so, the candidate connected domain area is filled;

A filling calculation unit, used to calculate the direction of the major axis of the ellipse circumscribed in the candidate connected region after filling, and obtain the angle between the direction and the set reference axis;

The angle threshold judgment unit is used to judge whether the angle is within a predetermined threshold range. If so, the candidate connected domain area is regarded as the area where the specific identification image is located; if not, the candidate connected domain area is discarded.

Preferably, the individual image feature screening unit includes:

A filling unit, configured to fill the candidate connected component region to obtain a filled image;

an edge contour acquiring unit, configured to acquire an edge contour image of the filled image by detecting the edge of the filled image;

a sampling unit, configured to respectively obtain sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image, and select the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range;

A sampling pixel point fitting unit is used to obtain sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image respectively, and select the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range;

The line segment combination judgment unit is used to judge whether the line segment combination information and intersection position information are the same as the line segment combination information and intersection position information of the standard specific identification image; if so, the candidate connected domain area is regarded as the specific identification image area; if not, the undetected specific identification image area is output.

Preferably, the sampling unit comprises:

a scanning unit, performing scanning based on the condition that there are two intersection points with the filling image, and using the intersection points as sampling pixel points;

The distance threshold judgment unit is used to judge whether the distance between the upper and lower edges and the distance between the left and right edges of the obtained sampling pixel points are within a set threshold range. If so, they are selected; if not, they are discarded.

Preferably, the threshold range is maxDist±maxDist×5%.

Preferably, it includes:

An affine transformation unit, configured to calculate an affine transformation matrix based on position information of a specific marker in a predetermined area or an image containing the predetermined area and position information of each intersection point;

The relevant information acquisition unit is used to acquire relevant information according to the affine transformation matrix.

Preferably, the specific identification image is a logo graphic of a bank card association or a bank card issuing organization.

Compared with the prior art, this application has the following advantages:

The present application obtains an image of a predetermined area as the target image to be inspected, binarizes the target image to be inspected, and calibrates the connected domains in the binarized image as candidate connected domain images. The connected domains that meet the characteristics of the specific identification image are screened out from these candidate connected domain images, and then it is determined that the specific identification information is detected.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flow chart of a method for detecting a specific identification image in a predetermined area provided by the present application;

FIG2 is a schematic diagram of virtual border calculation in a method for detecting a specific identification image in a predetermined area provided by the present application;

FIG3 is a schematic diagram of a virtual edge in a method for detecting a specific identification image in a predetermined area provided by the present application;

FIG4 is a schematic diagram of a first embodiment of a method for detecting a specific identification image in a predetermined area provided by the present application for intercepting an area containing a specific identification image;

FIG5 is a flowchart of a second embodiment of a method for detecting a specific identification image in a predetermined area provided by the present application for intercepting an area containing a specific identification image;

FIG6 is a schematic diagram of a binary image after a target image to be detected is binarized in a method for detecting a specific identification image in a predetermined area provided by the present application;

FIG7 is a flowchart of a method for detecting a specific identification image in a predetermined area provided by the present application, for performing a first screening of candidate connected domain regions;

FIG8 is a flowchart of a second screening of candidate connected areas in a method for detecting a specific identification image in a predetermined area provided by the present application;

FIG9 is a schematic diagram of detecting the upper and lower placement of the UnionPay logo in a method for detecting a specific logo image in a predetermined area provided by the present application;

FIG10 is a block diagram of the structure of a device for detecting a specific identification image in a predetermined area provided by the present application.

DETAILED DESCRIPTION

The following description sets forth many specific details to facilitate a thorough understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the scope of the present application. Therefore, the present application is not limited to the specific implementations disclosed below.

This application takes the area where the bank card surface is located as the predetermined area and the UnionPay logo on the bank card as the specific logo to illustrate the specific implementation process of this application. However, it should be noted that the method of this application is not limited to the recognition of the UnionPay logo on the bank card. The method of this application can be used for all recognition scenarios with specific known logos in the predetermined area. The specific known logo can be a product logo, a corporate logo, etc. The predetermined area of the product logo can be the product itself; the predetermined area of the corporate logo can be any carrier itself containing the logo.

Please refer to FIG1 , which is a flow chart of a first embodiment of a method for detecting a specific identification image in a predetermined area provided by the present application. In this embodiment, the method includes the following steps:

Step S100: Acquire an image of a predetermined area as an image of a target to be inspected;

Step S200: performing binarization processing on the target image to be inspected to obtain a binarized image and an inverse image of the binarized image, both of which are referred to as the target binary image to be inspected;

Step S300: calibrating the connected domain of the binary image of the target to be detected, and taking the connected domain area where the number of pixels in the calibrated connected domain meets the set threshold range as the candidate connected domain area;

Step S400: Acquire the image features of the candidate connected domain regions one by one, and compare the image features of the candidate connected domain regions with the corresponding image features of the standard specific identification image, and regard the candidate connected domain regions with the same or closest comparison results as the locations of the specific identification images.

Through the above steps, when scanning and analyzing the data, it is performed in a smaller area, so the above detection method can reduce the amount of calculation in the detection process, reduce the calculation complexity, and can quickly complete the detection based on feature comparison.

The following describes the specific implementation process of the above steps:

The step S100: obtaining an image of a predetermined area as an image of a target to be inspected.

The predetermined area in this step can be the bank card surface area. In other words, an image of the bank card surface or an image containing the bank card surface area can be obtained. This can be obtained manually or automatically using a mobile terminal, for example, by using a mobile terminal's camera to capture the predetermined area or an image containing the predetermined area. Selecting the predetermined area can reduce the detection scanning range, thereby increasing detection speed and reducing computational complexity.

To further improve detection efficiency and reduce detection of non-UnionPay logo areas, the present application can, when acquiring the bank card image or the image containing the bank card area, first obtain the bank card image or the preview image containing the bank card face area image, and based on the actual size of the standard bank card and the resolution of the preview image, obtain the length and width of the virtual frame that frames the preview image; then calculate the upper, lower, left, and right boundaries of the virtual frame based on the length and width of the virtual frame, as follows:

Refer to FIG2 , which is a schematic diagram of virtual bounding box calculation in a method for detecting a specific identification image in a predetermined area provided by the present application. According to the "Card BIN Number and Identification Rules" in Volume 3 of the "UnionPay Card Business Operation Regulations", the standard size parameters of bank cards can be obtained. The card width is 85.60mm±0.30mm, and the card height is 53.98mm±0.30mm. In this application, the card error range of 0.30mm is ignored, that is, the card width CardWidthGdt = 85.6, and the card height CardHeightGdt = 53.98; assuming that the screen resolution of the mobile terminal is length × height, that is, ScreenWidth × PreviewHeight; the resolution parameter of the preview image is width × height, that is, PreviewWidth × PreviewHeight; the ratio parameter of the virtual border to the preview image is DisplayRatio; when calculating the virtual border, first determine whether PreviewHeight (preview image height) × CardWidthGdt (card width) is greater than (not equal to) PreviewWidth (preview image width) × CardHeightGdt (card height). If so, then:

ScaledWidth＝CardWidthGdt×PreviewHeight/CardHeightGdt×DisplayRatio;

ScaledHeight＝PreviewHeight×DisplayRatio;

If not, then:

ScaledHeight＝CardHeightGdt×PreviewWidth/CardWidthGdt×DisplayRatio;

ScaledWidth＝PreviewWidth×DisplayRatio;

After the width and height of the virtual frame are calculated, the upper, lower, left, and right boundaries of the virtual frame can be determined according to the resolution parameters of the mobile terminal screen, as follows:

Top = ScreenHeight / 2 - ScaledHeight / 2;

Bottom border = ScreenHeight/2 + ScaledHeight/2;

Left border Left = ScreenWidth/2 - ScaledWidth/2;

Right border Right = ScreenWidth/2 + ScaledWidth/2.

Refer to FIG3 , which is a schematic diagram of a virtual edge in a method for detecting a specific identification image in a predetermined area provided by the present application.

The virtual border covers the bank card image or the area containing the bank card image, and the area within the virtual border is used as the target image to be inspected, thereby reducing the detection range. Alternatively, within the coverage of the virtual border, the area containing the UnionPay logo image can be intercepted based on the specific position information of the UnionPay logo in a standard bank, and this area is used as the target image to be inspected, thereby further minimizing the detection range, improving detection speed, and reducing computational complexity.

This application describes two implementation methods for capturing the area containing the UnionPay logo image.

Implementation method 1:

Referring to Figure 4, which is a schematic diagram illustrating a first embodiment of a method for detecting a specific logo image in a predetermined area provided by this application, for intercepting an area containing a specific logo image, it is known that the location of the UnionPay logo on a bank card is predetermined. The UnionPay logo area is located on the right side of the front of the card and is available in either a bottom-mounted or top-mounted format.

When the UnionPay logo is placed in the bottom format, it must be located in the lower right corner of the card. The specific requirements are: the width of the UnionPay logo is 22mm, the height is 15mm, the distance from the right edge of the UnionPay logo to the right edge of the card is 2mm, and the distance from the bottom edge of the UnionPay logo to the bottom edge of the card is 2mm.

When the UnionPay logo is placed in the top format, it must be located in the upper right corner of the card. The specific requirements are: the width of the UnionPay logo is 22mm, the height is 15mm, the distance from the right edge of the UnionPay logo to the right edge of the card is 2mm, and the distance from the top edge of the UnionPay logo to the top edge of the card is 2mm.

According to the above position of the UnionPay logo, when capturing a bank card image, only the upper right corner and lower right corner of the bank card image or the image containing the bank card area can be processed. For example, assuming that the resolution (width × height) of the bank card image or the image containing the bank card area is: Width × Height, the images that need to be captured are [Width/2, Width] × [0, Height] and [Width/2, Width] × [Height/2, Height] respectively.

Implementation method 2:

Please refer to FIG. 5 , which provides a flow chart of a second embodiment of a method for detecting a specific identification image in a predetermined area and capturing an area containing a specific identification image; the method includes:

Step S101: creating a standard image template for a bank card image;

Step S102: Traversing the standard image template according to the bank card image or the image containing the bank card area, and obtaining position information of the UnionPay logo image in the bank card image or the image containing the bank card area based on the correspondence between the standard image template and the bank card image or the image containing the bank card area;

Step S103: intercepting the region containing the UnionPay logo image according to the position information, and using the intercepted region containing the UnionPay logo image as the target image to be inspected.

It is understandable that the present application can also directly intercept the image area containing the UnionPay logo based on the bank card image or the image containing the bank card area, according to the specific position information of the UnionPay logo or the standard image template of the bank card, and use the image area as the target image to be inspected; it can also intercept only the part of the image area containing the UnionPay logo within the virtual border based on the virtual border, and discard the image information outside the virtual border, thereby improving the efficiency of detection.

The step S200: performing binarization processing on the target image to be detected to obtain a binarized image and an inverse image of the binarized image, both of which are referred to as the target binary image to be detected.

Referring to FIG6 , FIG6 is a schematic diagram of a binary image after a target image to be detected is binarized in a method for detecting a specific identification image in a predetermined area provided by the present application;

In this step, the target image to be inspected, that is, the image of the area containing the UnionPay logo image, is binarized. Image binarization refers to setting the grayscale value of the pixels on the image to 0 or 255, that is, making the entire image appear to be only black and white; that is, the grayscale image with 256 brightness levels is selected through an appropriate threshold to obtain a binary image that can still reflect the overall and local features of the image. To process and analyze binary images, the original image must first be grayscaled, and then the grayscale image is binarized to obtain a binary image. This is beneficial for further processing of the image. The collective properties of the image are only related to the positions of the points with pixel values of 0 or 255, and no longer involve the multi-level values of the pixels, making the processing simpler and reducing the amount of data processing and compression.

In this step, the threshold selection during the binarization process can use the Otsu method to segment the grayscale image into two parts: background and target (foreground). The Otsu method, also known as the maximum inter-class variance method, is an adaptive threshold determination method that divides the image into two parts, background and target, according to the grayscale characteristics of the image. The larger the inter-class variance between the background and target, the greater the difference between the background and target in the image. The variance of the foreground and background images is calculated by the proportion of foreground pixels in the image and the average grayscale value; the proportion of background pixels in the image and the average grayscale value. The grayscale value corresponding to the maximum variance is selected as the global threshold, and the image is binarized according to this global threshold.

Since the white pixels (target pixels) in the binary image are to be marked during the analysis of the binary image, and the image containing the UnionPay logo is binarized as the target image to be detected, the UnionPay logo may be binarized as foreground (1) or background (0), therefore, after obtaining the binary image, the binary image can be inverted, and both the binary image and the inverted image are used as the target binary image to be detected.

The most important method of binary image analysis is connected region labeling, which is the basis of all binary image analysis. It marks the white pixels (targets) in the binary image so that each individual connected region forms a marked block. Further, we can obtain the geometric parameters of these blocks, such as the outline, circumscribed rectangle, center of mass, invariant moment, etc.

The connected component labeling process for binary images involves extracting adjacent (4- or 8-adjacent) sets of pixels with a value of "1" from a dot matrix composed of white pixels (typically represented by "1" in binary images) and black pixels (typically represented by "0"). Binary images contain two colors: black (pixel value 0) and white (pixel value 1 or 255), which serve as the target color and background color, respectively. The labeling algorithm only labels the target pixels.

The step S300 is: calibrating the connected domain of the binary image of the target to be detected, and taking the connected domain in which the number of pixels in the calibrated connected domain meets a set threshold range as a candidate connected domain.

Connected domain calibration is the basis of all binary image analysis. The connected domain marking processing operation of a binary image is to extract the adjacent pixel sets with pixel value "1" from a raster image composed of white pixels (usually represented by "1" in binary images) and black pixels (usually represented by "0"). In other words, the binary image contains two colors, black (pixel value 0) and white (pixel value 1 or 255), which serve as the target color and background color respectively. When calibrating the connected domain, the calibration algorithm only calibrates the target pixels (white pixels).

In the connected component demarcation algorithm, there are two common adjacency relationships: 4-adjacency and 8-adjacency. 4-adjacency consists of four points: up, down, left, and right. 8-adjacency consists of eight points, including diagonal points: up, down, left, right, upper-left, and upper-right. A connected component is defined as a set of points that are connected to each other in the order of up, down, left, and right, or up, down, left, right, upper-left, and upper-right.

In this step, the captured area containing the UnionPay logo image is binarized, and the connected domain calibration method described above is used on the binary image to calibrate the connected domains in the binary image. In other words, all connected domains with a pixel value of 1 are calibrated. A threshold value is determined for each independently calibrated connected domain. If the number of pixels in the connected domain is within a given threshold range, the connected domain is retained. Otherwise, a connected domain calibration is performed on the inverted image of the target binary image to be inspected, and a threshold value is determined for the number of pixels in the calibrated connected domain in the inverted image. If the number of pixels in the connected domain is within the threshold range, the connected domain is retained; otherwise, the connected domain is discarded.

It should be noted that, for the connected domain formed by the UnionPay logo image, the number of pixels in the connected domain will be within a reasonable range, and the reasonable range can be used to set the threshold interval of the connected domain of the UnionPay logo image.

The specific calculation of the threshold interval of the above UnionPay logo connectivity domain can be referred to as follows:

LowThreshhold = Width × Height × lower limit coefficient;

HighThreshhold = Width × Height × upper limit coefficient;

Among them, Width and Height are the width and height of the acquired predetermined area image or the image containing the predetermined area, that is, Width and Height can be the width and height of the target image to be inspected, and the lower limit coefficient and upper limit values can be empirical values, which are 0.011 and 0.043 respectively.

When the detected specific identification information is not the UnionPay logo, a threshold interval may also be set according to a reasonable range of the number of pixels in the connected domain of the specific logo.

The step S400: obtaining the image features of the candidate connected domain regions one by one, and comparing the image features of the candidate connected domain regions with the corresponding image features of the standard specific identification image, and considering the candidate connected domain regions with the same or similar comparison structures as the locations of the specific identification image.

The specific implementation process of this step is to compare the corresponding image features of the candidate UnionPay logo connected domain area and the standard UnionPay logo image. Specifically, a single image feature comparison or a combined image feature comparison can be used; wherein, the image features include: the volume ratio of the connected domain, the major axis direction of the ellipse circumscribed in the connected domain, the outer contour shape of the connected domain and/or the aspect ratio of the rectangle circumscribed in the connected domain, etc. It can also be understood that the above features belong to the shape features of the image, and the shape features can also include distance features, image convexity features, length features, etc.

In this step, the volume ratio of the connected domain and the major axis direction of the ellipse circumscribed by the connected domain may be used as the first combined image feature to screen the connected domain area of the candidate UnionPay logo image.

Please refer to FIG. 7 , which is a flowchart of performing a first screening of candidate connected domain regions in a method for detecting a specific identification image in a predetermined region provided by the present application.

The specific steps include:

Step S701: Obtain the bounding box of the candidate connected region;

In this step, the image of the connected domain of the candidate UnionPay logo can be scanned inward from the outermost edge. After the foreground color (white pixel) is detected, the scanning is stopped to obtain an outer bounding box, which is a rectangle.

Step S702: Calculating the volume ratio of the pixels of the foreground color in the candidate connected domain within the bounding box to the area of the bounding box;

In this step, the volume ratio of the connected domain can be obtained by counting the number of foreground color (white) pixels within the rectangular outer frame and dividing the number of white pixels by the area of the rectangular outer frame.

Step S703: determining whether the volume ratio is greater than a preset threshold range of a specific identification image; if not, discarding the candidate connected domain area; if so, filling the candidate connected domain area;

In this step, a predetermined threshold value Tsolid for the volume ratio of the connected domain of the standard UnionPay logo can be used to determine whether the volume ratio of the connected domain of the candidate UnionPay logo matches the image characteristics of the standard connected domain of the UnionPay logo. The threshold value Tsolid can be obtained using information about the standard UnionPay logo image. For example, the standard UnionPay logo image can be obtained from various angles and its volume ratio to the external matrix can be calculated to select the threshold value Tsolid. The volume ratio of the connected domain of the candidate UnionPay logo can then be compared with Tsolid to determine whether the candidate connected domain is a UnionPay logo image.

When the volume ratio of the candidate UnionPay logo connected domain is greater than the threshold value Tsolid, the candidate UnionPay logo connected domain is considered a UnionPay logo connected domain. Because the candidate UnionPay logo connected domain may contain non-UnionPay logo connected domains with the same image characteristics as the standard UnionPay logo connected domain volume ratio, the present application further determines the candidate UnionPay logo by combining another image feature, namely, the image feature along the major axis of the connected domain ellipse. After determining that the volume ratio of the candidate UnionPay logo connected domain is greater than the threshold value Tsolid, the process proceeds to determining the image feature along the major axis of the connected domain ellipse, as described in step S704.

Step S704: filling the candidate connected domain area;

In this step, the UnionPay logo candidate connected domain area is filled with the foreground color (white).

Step S705: Calculate the direction of the filled UnionPay logo candidate connected domain area;

In this step, the direction of the connected component can be calculated by using ellipse fitting. In this embodiment, the direction of the connected component is the direction of the major axis of the circumscribed ellipse obtained on the connected component. The ellipse fitting is performed by performing edge detection on the candidate connected component, recording the points on the edge, and performing ellipse fitting on these points.

After calculating the direction of the major axis of the ellipse fit, a coordinate axis is established with the lower left corner of the connected domain as the origin. The X-axis of this coordinate axis is defined as the reference axis, and the angle formed between the major axis of the ellipse and the X-axis is then calculated. Similarly, for the standard UnionPay logo image, the angle formed between the major axis of the ellipse and the X-axis can also be calculated.

Step S706: determining whether the angle is within a predetermined threshold range; if so, considering the candidate connected domain area of the UnionPay logo as the area where the UnionPay logo image is located; if not, discarding the candidate connected domain area of the UnionPay logo;

In this step, the threshold range of the angle can be based on the above-mentioned ellipse fitting of the standard UnionPay logo image and the calculation of the angle between the major axis of the ellipse and the reference axis. Because, when obtaining the target image of the UnionPay logo to be inspected, there may be factors such as the acquisition angle, which will also affect the angle of the candidate connected domain image. Therefore, this application sets the threshold range of the angle based on the angle between the major axis of the ellipse in the standard UnionPay logo and the reference axis to avoid affecting the judgment due to the problem of the acquisition angle. The reference axis here is also the X-axis. The setting of the above-mentioned reference axis can be set according to the actual judgment needs, that is, it can also be the Y-axis, which can meet the requirements for judging the major axis direction characteristics of the ellipse circumscribed in the connected domain.

It should be noted that when a specific logo has two distinct circumscribed edges, namely, an inner edge and an outer edge connected domain, a secondary volume ratio determination can be performed on the UnionPay logo candidate connected domain image based on the above. However, the threshold used in this secondary volume ratio determination is not the aforementioned Tsolid, but a different threshold Tsolid' set specifically for this determination. The detection and determination process is the same as steps S701-S706 above. It should be noted that the pixel values of the inner connected domain are different from those of the outer connected domain, so different thresholds are used.

After the above steps, it can be determined that the UnionPay logo candidate connected domain image is the UnionPay logo image. To further confirm, this application judges that the angle between the major axis of the ellipse and the reference axis is within a predetermined threshold range based on the above steps, and then uses the outer contour shape of the connected domain as a separate condition to perform a second screening of the candidate connected domain area.

8 , which is a flowchart of a method provided by the present application for detecting a specific identification image in a predetermined area, in which the outer contour shape of the connected domain is used as a separate condition for performing a second screening on the candidate connected domain area.

The specific steps include:

Step S801: Filling the candidate connected component area to obtain a filled image;

This step may use the filling method in step S704 to fill the candidate connected component area.

Step S802: detecting the edge of the filled image to obtain an edge contour image of the filled image;

In this step, after the candidate connected domain area is filled with white, the white area becomes the candidate connected domain area. Edge detection methods are then used to obtain an edge contour image of the filled image. Edge detection methods are commonly used in image processing and can capture important features of objects in an image. Common edge detection methods include: edge detection, gradient edge detection, Roberts edge detection operator, Sobel edge detection operator, Prewitt edge detection operator, and Laplace edge detection operator. In this application, the Sobel edge detection operator is used to detect the edges of the filled image. The Sobel edge detection operator primarily detects edges based on the weighted grayscale difference between the upper and lower and left and right neighboring pixels of a pixel, reaching an extreme value at the edge. It has a noise-smoothing effect and can provide relatively accurate edge direction information. There are two Sobel operators: one for detecting horizontal edges and the other for detecting vertical edges. Compared to the Prewitt operator, the Sobel operator weights the influence of pixel position, which can reduce edge blur, resulting in better results. Of course, other edge detection operators can also be used to detect the edges of the filled image according to different needs.

Step S803: acquiring sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image respectively, and selecting the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range;

In this step, a scanning method is used to sample pixel points at the upper and lower edges and the left and right edges of the edge contour image. In the case of the UnionPay logo as an example, the edge contour of the fill image is a rectangular image. The upper and lower edges and the left and right edges of the rectangular image are scanned separately. The scanning method for the upper and lower edges and the left and right edges can be used in the same way. The upper and lower edges are used as an example below.

Scan the edges of the rectangular image column by column, select pixels where the scan line intersects the edge image, calculate the distances between these pixels, and count the pixels with the maximum value (maxDist) between all pixels as sampling pixels. Because the edges of rectangular images may have jagged edges, a threshold range of the maximum distance is set to select sampling pixels. The threshold range can be set to maxDist±maxDist×5%. Pixels within this threshold range are selected as sampling pixels, and pixels outside this threshold range are discarded. This results in sampling pixels for the upper and lower edges of the rectangular image. Similarly, sampling pixels for the left and right edges can be obtained.

Step S804: fitting the selected sampling pixel points to obtain line segment combination information of the edge contour image.

In step S804, a polynomial curve fitting method can be used to fit the sampled pixel points of the four sides of the candidate connected domain rectangular image selected in step S803, thereby obtaining four line segments, and four vertices can be obtained by combining the four line segments. The specific implementation method of step S804 is illustrated by taking the UnionPay logo of a bank card as an example. It can be understood that if other logos can use different fitting methods or calculation methods to fit the sampled pixel points.

Step S805: determining whether the line segment combination information is the same as the line segment combination information of the standard specific identification image; if so, treating the candidate connected domain area as the specific identification image area; if not, outputting an undetected specific identification image area.

In this step, for the line segment combination obtained in step S804, that is, four line segments and four vertices obtained based on the combination of the four line segments, the four line segments and four vertices are compared with the image of the standard bank card UnionPay logo. If they are the same, the UnionPay logo candidate connected domain area is the UnionPay logo image.

It should be noted that after the above steps, the candidate UnionPay logo is detected, and the system can output the four corner points of the UnionPay logo (see Figure 9). At the same time, if the combination of information detected, for example: more or less than 4 vertex information, it can be determined that the candidate area is not a UnionPay icon. If the detected UnionPay logo candidate area is empty, the user is prompted that no UnionPay logo is detected. If the number of detected UnionPay logo candidate areas is more than one, all candidate areas are provided to the user for self-selection.

The above is an explanation of detecting specific logos using the UnionPay logo as an example. For the UnionPay logo, its image features are shape features, including attributes such as a parallelogram and four vertices; when the specific logo is text or a product logo or a corporate logo, etc., it can be detected by the image features of this application, for example: the logos of various issuing banks. Taking the logo of China Merchants Bank as an example, please refer to Figure 3. The image features of the China Merchants Bank logo image can be a circle, the letter M and the vertices of the letter M, a triangle or a concave and convex attribute, etc. The above detection method can be used to detect the logo through these feature attributes.

It can be seen that the method for detecting a specific logo provided in this application is not limited to the above-mentioned UnionPay logo.

After the UnionPay logo is detected, the affine transformation matrix of the UnionPay logo image can be calculated based on the four vertex information of the detected UnionPay logo and the vertex information of the standard UnionPay information; based on this matrix, other information on the bank card can be calculated. The affine transformation matrix can be obtained as follows:

Assuming the coordinates of the four vertices of the standard UnionPay logo are [x1, y1], [x2, y2], [x3, y3], and [x4, y4], and the coordinates of the four vertices in the detected candidate connected domain image are [X1, Y1], [X2, Y2], [X3, Y3], and [X4, Y4], the affine transformation (projective) matrix M can be calculated based on the correspondence between the four points. For each pixel in the detected image, the image after the projective transformation can be calculated. The visual field of this image is the orthographic image. According to the regulations on the bank card, any information on the bank card can be obtained. For example, if the card number on the bank card is needed, the pixel of the card number is multiplied by the matrix M to obtain the orthographic image of the card number, thereby accurately obtaining the card number information.

The above discloses a method embodiment of the present application for detecting a specific identification image in a predetermined area. Corresponding to the method embodiment, the present application also discloses an apparatus embodiment for detecting a specific identification image in a predetermined area. Please refer to Figure 10, which is a schematic diagram of the structure of an embodiment of the apparatus for detecting a specific identification image in a predetermined area. Because the apparatus embodiment is substantially similar to the method embodiment, the description is relatively simple. For relevant details, please refer to the description of the method embodiment. The apparatus embodiment described below is merely illustrative.

As shown in FIG10 , the present application provides a device for detecting a specific identification image in a predetermined area, comprising:

An acquisition unit A100 is configured to acquire the predetermined area or an image containing the predetermined area as an image of an object to be inspected;

A binarization processing unit A200 is used to perform binarization processing on the target image to obtain a binarized image and an inverse image of the binarized image, both of which are referred to as the target binary image;

The connected domain calibration unit A300 is used to calibrate the connected domain of the binary image of the target to be detected, and select the connected domain whose number of pixels in the calibrated connected domain meets the set threshold range as the candidate connected domain;

The image feature comparison unit A400 is used to obtain the image features of the candidate connected domain one by one, and compare the image features of the candidate connected domain with the corresponding image features of the standard specific identification image, and regard the candidate connected domain with the same or closest comparison results as the location of the specific identification image.

The acquisition unit A100 includes:

a preview image acquiring unit, configured to acquire a preview image of the predetermined area or an image containing the predetermined area;

A calculation unit is used to calculate the area range of the preview image; the calculation unit is to obtain virtual frame parameters for framing the preview image based on the actual size parameters of the predetermined area and the resolution parameters of the preview image, specifically:

Determining, based on the actual size of the predetermined area or the image containing the predetermined area, whether a value obtained by multiplying the height of the preview image by the actual width of the predetermined area or the image containing the predetermined area is greater than a value obtained by multiplying the width of the preview image by the actual height of the predetermined area or the image containing the predetermined area;

If so, the width of the virtual border is:

ScaledWidth＝CardWidthGdt×PreviewHeight/CardHeightGdt×DisplayRatio;

Wherein, CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; PreviewHeight is the height of the preview image; CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image;

The height of the virtual border is:

ScaledHeight＝PreviewHeight×DisplayRatio;

Where PreviewHeight is the height of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image;

If not, the width of the virtual border is:

ScaledWidth＝PreviewWidth×DisplayRatio;

Where PreviewWidth is the width of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image;

The height of the virtual border is:

ScaledHeight＝CardHeightGdt×PreviewWidth/CardWidthGdt×DisplayRatio;

Wherein, CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; PreviewWidth is the width of the preview image; CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image;

The upper boundary of the virtual border is half of the difference between the display screen height and the virtual border height; the lower boundary of the virtual border is half of the sum of the display screen height and the virtual border height; the left boundary of the virtual border is half of the difference between the display screen width and the virtual border width; and the right boundary of the virtual border is half of the sum of the display screen width and the virtual border width.

The framing unit is used to frame the preview image within the area, obtain the predetermined area or an image containing the predetermined area, and use the image within the framed area as the target image to be inspected.

The acquisition unit A100 further includes:

The interception unit is used to intercept the area range containing the specific identification image according to the specific position information of the specific identification image in the predetermined area or the predetermined area image, as the target image to be inspected. The specific position information in the interception unit is the pre-set position of the specific identification image in the predetermined area or the image containing the predetermined area. In the case where the UnionPay logo of the bank card is used as the specific identification, the specific position can be known according to the records in the "Card BIN Number and Identification Rules" in Volume 3 of the "UnionPay Card Business Operation Regulations". The specific position of the UnionPay logo is located in the upper right corner or lower right corner of the bank card surface area. It is understandable that when the specific identification information is not the UnionPay logo, it can also be located in the upper left corner or lower left corner, etc.

The interception unit can also be implemented in the following ways:

An establishing unit, configured to establish a predetermined area or a standard image template containing the predetermined area;

a traversal unit, configured to traverse the standard image template according to the predetermined area or the image containing the predetermined area, and obtain position information of the specific identification image in the predetermined area or the image containing the predetermined area through a correspondence between the standard image template and the predetermined area or the image containing the predetermined area;

The target image to be inspected determining unit is configured to intercept the predetermined area or an image area containing a specific mark in the image containing the predetermined area according to the position information as the target image to be inspected.

The upper limit value of the threshold interval in the connected domain calibration unit A300 is the product of the width, height and upper limit coefficient of the predetermined area or the image containing the predetermined area; the lower limit value is the product of the width, height and lower limit coefficient of the predetermined area or the image containing the predetermined area.

The image feature comparison unit A400 includes a combined image feature screening unit or a separate image feature screening unit, by adopting separate image feature comparison or combined image feature comparison; wherein, the image features include the volume ratio of the connected domain, the major axis direction of the ellipse circumscribed in the connected domain, the outer contour shape of the connected domain and/or the aspect ratio of the rectangle circumscribed in the connected domain.

Combined image feature screening unit, specifically:

An outer bounding box obtaining unit, used to obtain the outer bounding box of the candidate connected domain area;

a volume ratio calculation unit, configured to calculate a volume ratio of pixels of the foreground color in the candidate connected domain within the circumscribed frame to the area of the circumscribed frame;

A volume ratio threshold judgment unit is used to judge whether the volume ratio is greater than a preset threshold range of a specific identification image; if not, the candidate connected domain area is discarded; if so, the candidate connected domain area is filled;

A filling calculation unit, used to calculate the direction of the major axis of the ellipse circumscribed in the candidate connected region after filling, and obtain the angle between the direction and the set reference axis;

The angle threshold judgment unit is used to judge whether the angle is within a predetermined threshold range. If so, the candidate connected domain area is regarded as the area where the specific identification image is located; if not, the candidate connected domain area is discarded.

Separate image feature screening unit, including:

A filling unit, configured to fill the candidate connected component region to obtain a filled image;

an edge contour acquiring unit, configured to acquire an edge contour image of the filled image by detecting the edge of the filled image;

a sampling unit, configured to respectively obtain sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image, and select the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range;

A sampling pixel point fitting unit is used to obtain sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image respectively, and select the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range;

The line segment combination judgment unit is used to judge whether the line segment combination information and intersection position information are the same as the line segment combination information and intersection position information of the standard specific identification image; if so, the candidate connected domain area is regarded as the specific identification image area; if not, the undetected specific identification image area is output.

The sampling unit comprises:

a scanning unit, performing scanning based on the condition that there are two intersection points with the filling image, and using the intersection points as sampling pixel points;

A distance threshold determination unit is configured to determine whether the distances between the upper and lower edges and the distances between the left and right edges of the sampled pixels are within a set threshold range. If so, the sampled pixels are selected; otherwise, the sampled pixels are discarded. In this distance threshold determination unit, the threshold range may be maxDist ± maxDist × 5%, where maxDist is the maximum distance between two sampled pixels.

The present application also includes an affine transformation unit for calculating an affine transformation matrix based on the position information of specific identifiers in a predetermined area or an image containing a predetermined area and the position information of each intersection; and a related information acquisition unit for acquiring related information based on the affine transformation matrix.

Although the present application is disclosed as above with the preferred embodiments, it is not intended to limit the present application. Any person skilled in the art may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and/or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

1. Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include non-transitory media such as modulated data signals and carrier waves.

2. Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

Claims (30)

1. A method for detecting a specific identification image in a predetermined area, comprising: Acquire an image of a predetermined area as an image of a target to be inspected; Binarizing the target image to be inspected to obtain a binary image and a negated image of the binary image, wherein both the binary image and the negated image are referred to as the binary image of the target image to be inspected; Calibrate the connected domain of the binary image of the target to be detected, and take the area where the number of pixels in the calibrated connected domain meets the set threshold range as the candidate connected domain area; The image features of the candidate connected domain areas are obtained one by one, and the image features of the candidate connected domain areas are compared with the corresponding image features of the standard specific identification image using a single image feature comparison or a combined image feature comparison, and the candidate connected domain areas with the same or closest comparison results are regarded as the locations of the specific identification images. 2. The method for detecting a specific identification image in a predetermined area according to claim 1, wherein obtaining the image of the predetermined area as the target image to be detected comprises: Acquiring a preview image of the predetermined area or an image containing the predetermined area; Calculating the area range of the preview image; The preview image is framed within the region, the predetermined region or an image containing the predetermined region is acquired, and the image within the framed region is used as a target image to be inspected. 3. The method for detecting a specific identification image in a predetermined area according to claim 2, wherein calculating the area range of the preview image comprises obtaining virtual frame parameters for framing the preview image based on actual size parameters of the predetermined area and resolution parameters of the preview image, specifically: determining, based on the actual size of the predetermined area or the image containing the predetermined area, whether a value obtained by multiplying the height of the preview image by the actual width of the predetermined area or the image containing the predetermined area is greater than a value obtained by multiplying the width of the preview image by the actual height of the predetermined area or the image containing the predetermined area; If so, the width of the virtual border is: ScaledWidth＝CardWidthGdt×PreviewHeight/CardHeightGdt×DisplayRatio; Wherein, CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; PreviewHeight is the height of the preview image; CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image; The height of the virtual border is: ScaledHeight＝PreviewHeight×DisplayRatio; Where PreviewHeight is the height of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image; If not, the width of the virtual border is: ScaledWidth＝PreviewWidth×DisplayRatio; Where PreviewWidth is the width of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image; The height of the virtual border is: ScaledHeight＝CardHeightGdt×PreviewWidth/CardWidthGdt×DisplayRatio; Wherein, CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; PreviewWidth is the width of the preview image; CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image; The upper boundary of the virtual border is half of the difference between the display screen height and the virtual border height; the lower boundary of the virtual border is half of the sum of the display screen height and the virtual border height; the left boundary of the virtual border is half of the difference between the display screen width and the virtual border width; and the right boundary of the virtual border is half of the sum of the display screen width and the virtual border width. 4. The method for detecting a specific identification image in a predetermined area according to any one of claims 1 to 3, wherein acquiring the predetermined area or an image containing the predetermined area as the target image to be detected comprises: According to the specific position information of the specific identification image in the predetermined area or the predetermined area image, the area range including the specific identification image is intercepted as the target image to be inspected. 5 . The method for detecting a specific identification image in a predetermined area according to claim 4 , wherein the specific position information is a pre-set position of the specific identification image in the predetermined area or in an image including the predetermined area. 6. The method for detecting a specific identification image in a predetermined area according to claim 5, wherein the specific position is the upper right corner, lower right corner, upper left corner, or lower left corner of the predetermined area or the image containing the predetermined area. 7. The method for detecting a specific identification image in a predetermined area according to claim 4, wherein, based on specific position information of the specific identification image in the predetermined area or the predetermined area image, the region containing the specific identification image is intercepted as the target image to be detected, comprising: Establishing a predetermined area or a standard image template containing a predetermined area; Traversing the standard image template according to the predetermined area or the image containing the predetermined area, and obtaining position information of the specific identification image in the predetermined area or the image containing the predetermined area through a correspondence between the standard image template and the predetermined area or the image containing the predetermined area; The predetermined area or the image area containing the specific mark in the image containing the predetermined area is intercepted according to the position information as the target image to be inspected. 8. The method for detecting a specific identification image in a predetermined area according to claim 1, wherein: The upper limit value of the threshold interval is the product of the width, height and upper limit coefficient of the acquired predetermined area image or the image containing the predetermined area; the lower limit value is the product of the width, height and lower limit coefficient of the acquired predetermined area image or the image containing the predetermined area. 9. The method for detecting a specific identification image in a predetermined area according to claim 1 is characterized in that the image features include the volume ratio of the connected domain, the major axis direction of the ellipse circumscribing the connected domain, the outer contour shape of the connected domain and/or the aspect ratio of the rectangle circumscribing the connected domain. 10. The method for detecting a specific identification image in a predetermined area according to claim 9, characterized in that the steps of acquiring image features of each candidate connected domain region one by one, comparing the image features of the candidate connected domain region with corresponding image features of the specific identification image, and considering the candidate connected domain region with the same or closest comparison results as the region where the specific identification image is located, comprise: The candidate connected domain regions are screened using the volume ratio of the connected domain and the major axis direction of the connected domain circumscribed ellipse as combined image features, specifically: Get the bounding box of the candidate connected region; Calculating a volume ratio of pixels of the foreground color in the candidate connected domain within the bounding box to the area of the bounding box; Determine whether the volume ratio is greater than a preset threshold range of a specific identification image; if not, discard the candidate connected domain area; if so, fill the candidate connected domain area; Calculating the direction of the major axis of the ellipse circumscribed in the filled candidate connected region, and obtaining the angle between the major axis and the set reference axis; It is determined whether the angle is within a predetermined threshold range. If so, the candidate connected domain area is regarded as the area where the specific identification image is located; if not, the candidate connected domain area is discarded. 11. The method for detecting a specific identification image in a predetermined area according to claim 10, further comprising: after determining that the angle between the direction and the reference axis is within a predetermined threshold range, further screening the candidate connected domain regions using the outer contour shape of the connected domain as a separate condition, specifically: Filling the candidate connected component area to obtain a filled image; Detecting the edge of the filled image to obtain an edge contour image of the filled image; Respectively obtaining sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image, and selecting the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range; Fitting the selected sampling pixel points to obtain line segment combination information of the edge contour image; Determine whether the line segment combination information and intersection position information are the same as the line segment combination information and intersection position information of the standard specific identification image; if so, regard the candidate connected domain area as the specific identification image area; if not, output the undetected specific identification image area. 12. The method for detecting a specific identification image in a predetermined area according to claim 11, wherein obtaining sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image comprises: Scanning is performed on the condition that there are two intersection points with the filling image, and the intersection points are used as sampling pixel points; The selecting of the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range includes: Determine whether the distances between the upper and lower edges and the distances between the left and right edges of the obtained sampling pixel points are within a set threshold range. If so, select them; if not, discard them. 13 . The method for detecting a specific identification image in a predetermined area according to claim 12 , wherein the threshold range is maxDist±maxDist×5%. 14. The method for detecting a specific identification image in a predetermined area according to claim 11, characterized by comprising: Calculating an affine transformation matrix based on the position information of a specific marker in a predetermined area or an image containing the predetermined area and the position information of each intersection point; Relevant information within the predetermined area is acquired according to the affine transformation matrix. 15. The method for detecting a specific identification image in a predetermined area according to any one of claims 1-3 and 5-14, wherein the specific identification image is a logo graphic of a bank card association or a bank card issuing organization. 16. A device for detecting a specific identification image in a predetermined area, comprising: an acquisition unit, configured to acquire an image of the predetermined area as an image of the target to be inspected; a binarization processing unit, configured to perform binarization processing on the target image to be detected to obtain a binarized image and a negated image of the binarized image, wherein both the binarized image and the negated image are referred to as the target binary image to be detected; A connected domain calibration unit is used to calibrate the connected domain of the binary image of the target to be detected, and to select the connected domain whose number of pixels in the calibrated connected domain meets the set threshold range as a candidate connected domain; An image feature comparison unit is used to obtain the image features of the candidate connected domain one by one, and compare the image features of the candidate connected domain with the corresponding image features of the standard specific identification image using a single image feature comparison or a combined image feature comparison, and regard the candidate connected domain with the same or closest comparison results as the location of the specific identification image. 17. The device for detecting a specific identification image in a predetermined area according to claim 16, wherein the acquiring unit comprises: a preview image acquiring unit, configured to acquire a preview image of the predetermined area or an image containing the predetermined area; A calculation unit, configured to calculate the area range of the preview image; The framing unit is used to frame the preview image within the area, obtain the predetermined area or an image containing the predetermined area, and use the image within the framed area as the target image to be inspected. 18. The device for detecting a specific identification image in a predetermined area according to claim 17, wherein the calculating unit is configured to obtain virtual frame parameters for framing the preview image based on actual size parameters of the predetermined area and resolution parameters of the preview image, specifically: determining, based on the actual size of the predetermined area or the image containing the predetermined area, whether a value obtained by multiplying the height of the preview image by the actual width of the predetermined area or the image containing the predetermined area is greater than a value obtained by multiplying the width of the preview image by the actual height of the predetermined area or the image containing the predetermined area; If so, the width of the virtual border is: ScaledWidth＝CardWidthGdt×PreviewHeight/CardHeightGdt×DisplayRatio; Wherein, CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; PreviewHeight is the height of the preview image; CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image; The height of the virtual border is: ScaledHeight＝PreviewHeight×DisplayRatio; Where PreviewHeight is the height of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image; If not, the width of the virtual border is: ScaledWidth = PreviewWidth × DisplayRatio; Where PreviewWidth is the width of the preview image, and DisplayRatio is the ratio of the virtual border to the preview image; The height of the virtual border is: ScaledHeight＝CardHeightGdt×PreviewWidth/CardWidthGdt×DisplayRatio; Wherein, CardHeightGdt is the actual height of the predetermined area or the image containing the predetermined area; PreviewWidth is the width of the preview image; CardWidthGdt is the actual width of the predetermined area or the image containing the predetermined area; DisplayRatio is the ratio of the virtual border to the preview image; The upper boundary of the virtual border is half of the difference between the display screen height and the virtual border height; the lower boundary of the virtual border is half of the sum of the display screen height and the virtual border height; the left boundary of the virtual border is half of the difference between the display screen width and the virtual border width; and the right boundary of the virtual border is half of the sum of the display screen width and the virtual border width. 19. The device for detecting a specific identification image in a predetermined area according to any one of claims 16 to 18, wherein the acquiring unit comprises: The interception unit is used to intercept the area range containing the specific identification image as the target image to be inspected according to the specific position information of the specific identification image in the predetermined area or the predetermined area image. 20. The device for detecting a specific identification image in a predetermined area according to claim 19, wherein the specific position information is a pre-set position of the specific identification image in the predetermined area or in an image including the predetermined area. 21. The device for detecting a specific identification image in a predetermined area according to claim 20, wherein the specific position is the upper right corner or lower right corner or the upper left corner or lower left corner of the predetermined area or the image containing the predetermined area. 22. The device for detecting a specific identification image in a predetermined area according to claim 19, wherein the interception unit comprises: An establishing unit, configured to establish a predetermined area or a standard image template containing the predetermined area; a traversal unit, configured to traverse the standard image template according to the predetermined area or the image containing the predetermined area, and obtain position information of the specific identification image in the predetermined area or the image containing the predetermined area through a correspondence between the standard image template and the predetermined area or the image containing the predetermined area; The target image to be inspected determining unit is configured to intercept the predetermined area or an image area containing a specific mark in the image containing the predetermined area according to the position information as the target image to be inspected. 23. The device for detecting a specific identification image in a predetermined area according to claim 16 is characterized in that the upper limit value of the threshold range is the product of the width, height and upper limit coefficient of the predetermined area or the image containing the predetermined area; the lower limit value is the product of the width, height and lower limit coefficient of the predetermined area or the image containing the predetermined area. 24. The device for detecting a specific identification image in a predetermined area according to claim 16, wherein the image features include the volume ratio of the connected domain, the direction of the major axis of the ellipse circumscribed by the connected domain, the shape of the outer contour of the connected domain and/or the aspect ratio of the rectangle circumscribed by the connected domain. 25. The device for detecting a specific identification image in a predetermined area according to claim 24, characterized in that it comprises: a combined image feature screening unit, specifically: An outer bounding box obtaining unit, used to obtain the outer bounding box of the candidate connected domain area; a volume ratio calculation unit, configured to calculate a volume ratio of pixels of the foreground color in the candidate connected domain within the circumscribed frame to the area of the circumscribed frame; A volume ratio threshold judgment unit is used to judge whether the volume ratio is greater than a preset threshold range of a specific identification image; if not, the candidate connected domain area is discarded; if so, the candidate connected domain area is filled; A filling calculation unit, used to calculate the direction of the major axis of the ellipse circumscribed in the candidate connected region after filling, and obtain the angle between the direction and the set reference axis; The angle threshold judgment unit is used to judge whether the angle is within a predetermined threshold range. If so, the candidate connected domain area is regarded as the area where the specific identification image is located; if not, the candidate connected domain area is discarded. 26. The device for detecting a specific identification image in a predetermined area according to claim 25, wherein the individual image feature screening unit comprises: A filling unit, configured to fill the candidate connected component region to obtain a filled image; an edge contour acquiring unit, configured to acquire an edge contour image of the filled image by detecting the edge of the filled image; a sampling unit, configured to respectively obtain sampling pixel points of the upper and lower edges and the left and right edges of the edge contour image, and select the upper and lower edge sampling pixel points and the left and right edge sampling pixel points within a set threshold range; A sampling pixel point fitting unit, used for fitting the selected sampling pixel points to obtain line segment combination information of the edge contour image; The line segment combination judgment unit is used to judge whether the line segment combination information and intersection position information are the same as the line segment combination information and intersection position information of the standard specific identification image; if so, the candidate connected domain area is regarded as the specific identification image area; if not, the undetected specific identification image area is output. 27. The device for detecting a specific identification image in a predetermined area according to claim 26, wherein the sampling unit comprises: a scanning unit, performing scanning based on the condition that there are two intersection points with the filling image, and using the intersection points as sampling pixel points; The distance threshold judgment unit is used to judge whether the distance between the upper and lower edges and the distance between the left and right edges of the obtained sampling pixel points are within a set threshold range. If so, they are selected; if not, they are discarded. 28. The device for detecting a specific identification image in a predetermined area according to claim 27, wherein the threshold range is maxDist±maxDist×5%. 29. The device for detecting a specific identification image in a predetermined area according to claim 26, characterized in that it comprises: An affine transformation unit, configured to calculate an affine transformation matrix based on position information of a specific marker in a predetermined area or an image containing the predetermined area and position information of each intersection point; The relevant information acquisition unit is used to acquire relevant information according to the affine transformation matrix. 30. The device for detecting a specific identification image in a predetermined area according to any one of claims 16-18 and 20-29, wherein the specific identification image is a logo graphic of a bank card association or a bank card issuing organization.