CN111027527A - Image feature identification method based on optical frequency features - Google Patents
Image feature identification method based on optical frequency features Download PDFInfo
- Publication number
- CN111027527A CN111027527A CN201911148466.5A CN201911148466A CN111027527A CN 111027527 A CN111027527 A CN 111027527A CN 201911148466 A CN201911148466 A CN 201911148466A CN 111027527 A CN111027527 A CN 111027527A
- Authority
- CN
- China
- Prior art keywords
- light source
- light
- image
- frequency
- optical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/141—Control of illumination
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/145—Illumination specially adapted for pattern recognition, e.g. using gratings
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an image characteristic identification method based on optical frequency characteristics, which can effectively improve the imaging quality of different substances in the same medium and improve the image identification rate and the detection rate. The image feature identification method based on the optical frequency features adopts an optical frequency feature image feature identification system; and comprises the steps of: s1, modulating the radiant energy flux by a single-frequency light source, S2, modulating the time of the single-frequency light source, S3, adjusting the irradiation position of the single light source, S4, irradiating by multiple-frequency light sources, and filtering out background noise by a light source with combined frequency; s5, forming specific optical characteristics for the target object based on the characteristics of the steps S1, S2, S3 and S4; and S6, automatically identifying the image features through a software algorithm. The method for identifying the image characteristics based on the optical frequency characteristics is used for identifying wide substances; the optical frequency can be freely selected.
Description
Technical Field
The invention relates to the field of optical characteristics, in particular to an image characteristic identification method based on optical frequency characteristics.
Background
It is well known that: the traditional optical imaging is based on a light source, a light transmission medium, an imaged object and an image surface; there are many types of light sources, but most utilize the visible light band to reflectively image objects. However, the absorption efficiency of different media to the same wavelength is different, and the reflected light wavelength is different: light absorption is a physical process by which light (electromagnetic radiation) passes through a material, interacting with the material, and the energy of the electromagnetic radiation is partially converted into other forms of energy. When the absorbed light energy is released in the form of heat energy, the photothermal conversion is formed; when the unabsorbed light energy is reflected, scattered or transmitted by an object, the color of the object we see is affected, and thus there is a difference in the imaging effect.
However, under the irradiation of the multispectral light source, light with different frequencies has different imaging quality, and the imaging quality, the identification effect and the detection result of the object are influenced due to different effects in the process of image gray level map or binarization processing. In the traditional imaging, the filter is optimized, but the working frequency range of the filter is limited, the commonly used filter wavelengths are 550nm, 590nm, 630nm, 690nm, 720nm, 760nm, 850nm and 950nm, the filter with partial frequency wavelengths is expensive to process and manufacture, the imaging mode is single, and the effective identification of the characteristics of various substances is difficult to realize.
The application of modulation of light intensity (radiant flux) is now substantially limited to the field of communications, facilitating the transmission of optical signals. At present, the frequency characteristics of optical intensity modulation are not utilized to perform characteristic identification on the image.
In various fields such as food, medicine, industry, military, scientific research and the like, a great variety of substances need to be rapidly identified and detected, and a great challenge is provided for the traditional imaging identification method. The image identification method based on the optical frequency characteristics modulates different substances with different optical frequency characteristics, different radiation fluxes, different irradiation time and different irradiation positions to form composite optical frequency characteristics, so that a plurality of image identification methods based on the optical frequency characteristics are generated, and the detection substances can be effectively identified.
Disclosure of Invention
The invention aims to solve the technical problem of providing an image feature identification method based on optical frequency features, which can effectively solve the defect of large imaging quality difference of different substances under the conditions of fixed frequency wavelength and fixed position light irradiation in conventional image identification, effectively improve the imaging quality of different substances in the same medium, improve the image identification rate and the detection rate and greatly reduce the calculation time of an image identification software algorithm.
The technical scheme adopted by the invention for solving the technical problems is as follows: an image characteristic identification method based on optical frequency characteristics adopts an optical frequency characteristic image characteristic identification system; the bubble identification system comprises a camera and a light source;
the light source comprises a first light source, a second light source and a third light source; the first light source is arranged on the first mounting plate, the second light source is arranged on the second mounting plate, and the third light source is arranged on the third mounting plate;
the first light source is controlled by a first industrial personal computer, the second light source is controlled by a second industrial personal computer, and the third light source is controlled by a third industrial personal computer;
the camera is connected with a PC controller; the light source is connected with an industrial personal computer; the light source is used for emitting a light source with a specific optical frequency, a specific light-emitting duration, a specific position and a specific shape; the light emitted by the light source is emitted to the object to be measured; the camera is used for shooting a measured object; the PC controller is used for collecting images and identifying the bubble space optical characteristics of the measured object;
and comprises the steps of:
s1, by modulating the radiant energy flux of the light source with single frequency, generating light with different energy amplitudes to illuminate the target object, generating images with different optical characteristics, and screening the characteristics of a series of images to achieve the purpose of identifying the target object; continuously modulating the radiant energy flux of a single light source to obtain continuously-changed analog quantity image data, and acquiring and identifying an optical characteristic image which is formed by a target object in a state of continuously modulating the radiant energy flux and continuously changes;
s2, controlling a light source switch to adjust the illumination time and the non-illumination time of the light source by modulating the time of the light source with single frequency so as to obtain images of the optical characteristics of the object with bright background and dark background under different illumination times;
s3, adjusting the irradiation position of a single light source through a light source adjusting support of a three-dimensional coordinate X/Y/Z, a pitch angle theta 1 and a rotation angle theta 2 in an adjustable three-dimensional space, obtaining optical characteristics of the same target in different irradiation directions at different irradiation directions and distances, and recognizing the target object through recognizing the characteristics;
s4, irradiating by using light sources with various frequencies, and according to the Grassmann color mixing law, the method is applicable to color light additive color mixing, namely light rays with different wavelengths are superposed, and the light and shade change is carried out on the color with the appointed wavelength through the superposition of the light with different wavelengths, so that the object with the appointed color is filtered and screened;
blue light (460nm) + yellow light (580nm) as white light
Red light (660nm) + cyan light (480nm) as white light
The imaging quality of different substances is different, and background noise is filtered by using a light source with combined frequency, so that the image contrast is enhanced, and the recognition rate is improved;
s5, based on the characteristics of the steps S1, S2, S3 and S4, performing combination matching on parameters such as radiation energy flux modulation, irradiation time modulation, distributed irradiation positions and the like of the multi-frequency light source to form structured light with a specific combination, and forming specific optical characteristics on the target object;
and S6, automatically identifying the image features through a software algorithm.
Preferably, the software algorithm in step S6 adopts an OSTU segmentation method;
the method specifically comprises the following steps: firstly, converting an image into a gray scale image; the number of pixels in the image, of which the gray value A is smaller than the threshold T, is H0, and the number of pixels of which the gray value A is larger than the threshold T is H1; recording the average gray value of H0 as H0 and the average gray value of H1 as H1 in the image with the image size of X X Y and the threshold value of T;
the probability that the pixel gray value is less than T is:
r0＝h0/(X*Y)；
the probability that the pixel gray value is greater than T is:
r1＝h1/(X*Y)；
h0+h1＝X*Y；
r0+r1＝1；
the average gray is multiplied by the probability and then added:
e＝r0*h0+r1*h1；
the between-class variance is:
d＝r0(h0-e)^2+r1(h1-e)^2；
d＝r0*r1(h0-h1)^2。
further, the software algorithm adopts a corrosion algorithm, an expansion algorithm or an expansion algorithm of gray-scale morphology.
The invention has the beneficial effects that: compared with the prior art, the image feature identification method based on the optical frequency features has the following advantages: the image characteristic identification method based on the optical frequency characteristics provides an image identification basis for material identification of related industries; different substances correspond to different response optimal frequencies, and the identification precision is high; the optimal frequency of the responsivity is used for identification, so that the identification speed is high; all substances have corresponding optical frequency characteristics, and the identified substances are wide; modulating light radiation flux and irradiation time, and adjusting multiple frequencies and multiple irradiation positions to form a composite identification method; the optical frequency can be freely selected.
Drawings
FIG. 1 is a schematic structural diagram of an optical frequency characteristic image feature identification system according to an embodiment of the present invention;
FIG. 2 is a schematic view of an embodiment of the invention for indoor irradiation of air bubbles in rainy or cloudy days;
FIG. 3 is an optical characteristic of an illumination bubble in a rainy day in an embodiment of the present invention;
FIG. 4 is a schematic view of a point source illuminating a bubble in an embodiment of the present invention;
FIG. 5 is an optical feature of a point source illuminating a bubble in an embodiment of the present invention;
FIG. 6 is a schematic diagram of two point light sources illuminating a bubble in an embodiment of the present invention;
FIG. 7 shows the optical characteristics of two point light sources illuminating a bubble in an embodiment of the present invention;
FIG. 8 is a schematic view of three point light sources illuminating a bubble in an embodiment of the present invention;
FIG. 9 shows the optical characteristics of a bubble illuminated by three point sources in an embodiment of the present invention;
FIG. 10 is a schematic diagram of two bar light sources illuminating a bubble in an embodiment of the present invention;
FIG. 11 is an optical signature of two bar light sources illuminating a bubble in an embodiment of the present invention;
FIG. 12 is a flow chart of a binary morphological erosion algorithm in accordance with an embodiment of the present invention;
FIG. 13 is a flow chart of a binary morphological dilation algorithm in accordance with an embodiment of the present invention;
FIG. 14 is a flowchart of a gray scale morphological erosion algorithm in an embodiment of the invention;
FIG. 15 is a flow chart of a gray scale morphological dilation algorithm in an embodiment of the present invention;
FIG. 16 is a photograph of a foreign object in an infusion bag in an embodiment of the invention;
FIG. 17 is the binarized image of FIG. 16 in an embodiment of the present invention;
FIG. 18 is a picture of the image of FIG. 17 after erosion expansion;
FIG. 19 is a picture of a foreign object detected in an embodiment of the present invention;
the following are marked in the figure: the system comprises a first industrial computer, a second industrial computer, a third industrial computer, a first light source, a second industrial computer, a second light source, a second industrial computer, a third light source, a third industrial computer, a first mounting plate, a second mounting plate, a third mounting plate, a first camera, a second camera, a 3-PC controller, a 4-first light source, a 5-first industrial computer, a 6-second light source, a 7-second industrial computer, a 8-third light source, a 9-third.
Detailed Description
The invention is further illustrated with reference to the following figures and examples.
As shown in fig. 1, the image feature identification method based on optical frequency features of the present invention employs an optical frequency feature image feature identification system; the bubble identification system comprises a camera 2 and a light source;
the light sources comprise a first light source 4, a second light source 6 and a third light source 8;
the first light source 4 is mounted on a first mounting plate 10, the second light source 6 is mounted on a second mounting plate 11, and the third light source 8 is mounted on a third mounting plate 12;
the first light source 4 is controlled by a first industrial personal computer 5, the second light source 6 is controlled by a second industrial personal computer 7, and the third light source 8 is controlled by a third industrial personal computer 9;
the camera 2 is connected with a PC controller 3; the light source is connected with an industrial personal computer; the light source is used for emitting a light source with a specific optical frequency, a specific light-emitting duration, a specific position and a specific shape; the light emitted by the light source is emitted to the object 1 to be measured; the camera 2 is used for shooting the measured object 1; the PC controller 3 is used for collecting images and identifying the bubble space optical characteristics of the object 1 to be measured;
and comprises the steps of:
s1, by modulating the radiant energy flux of the light source with single frequency, generating light with different energy amplitudes to illuminate the target object, generating images with different optical characteristics, and screening the characteristics of a series of images to achieve the purpose of identifying the target object; continuously modulating the radiant energy flux of a single light source to obtain continuously-changed analog quantity image data, and acquiring and identifying an optical characteristic image which is formed by a target object in a state of continuously modulating the radiant energy flux and continuously changes;
s2, controlling a light source switch to adjust the illumination time and the non-illumination time of the light source by modulating the time of the light source with single frequency so as to obtain images of the optical characteristics of the object with bright background and dark background under different illumination times;
s3, adjusting the irradiation position of a single light source through a light source adjusting support of a three-dimensional coordinate X/Y/Z, a pitch angle theta 1 and a rotation angle theta 2 in an adjustable three-dimensional space, obtaining optical characteristics of the same target in different irradiation directions at different irradiation directions and distances, and recognizing the target object through recognizing the characteristics;
s4, irradiating by using light sources with various frequencies, and according to the Grassmann color mixing law, the method is applicable to color light additive color mixing, namely light rays with different wavelengths are superposed, and the light and shade change is carried out on the color with the appointed wavelength through the superposition of the light with different wavelengths, so that the object with the appointed color is filtered and screened;
blue light (460nm) + yellow light (580nm) as white light
Red light (660nm) + cyan light (480nm) as white light
The imaging quality of different substances is different, and background noise is filtered by using a light source with combined frequency, so that the image contrast is enhanced, and the recognition rate is improved;
s5, based on the characteristics of the steps S1, S2, S3 and S4, performing combination matching on parameters such as radiation energy flux modulation, irradiation time modulation, distributed irradiation positions and the like of the multi-frequency light source to form structured light with a specific combination, and forming specific optical characteristics on the target object;
and S6, automatically identifying the image features through a software algorithm.
Specifically, the software algorithm in step S6 adopts an OSTU segmentation method; the method specifically comprises the following steps: firstly, converting an image into a gray scale image; the number of pixels in the image, of which the gray value A is smaller than the threshold T, is H0, and the number of pixels of which the gray value A is larger than the threshold T is H1; recording the average gray value of H0 as H0 and the average gray value of H1 as H1 in the image with the image size of X X Y and the threshold value of T;
the probability that the pixel gray value is less than T is:
r0＝h0/(X*Y)；
the probability that the pixel gray value is greater than T is:
r1＝h1/(X*Y)；
h0+h1＝X*Y；
r0+r1＝1；
the average gray is multiplied by the probability and then added:
e＝r0*h0+r1*h1；
the between-class variance is:
d＝r0(h0-e)^2+r1(h1-e)^2；
d＝r0*r1(h0-h1)^2。
preferably, the software algorithm adopts an erosion algorithm, an expansion algorithm or an expansion algorithm of gray scale morphology.
OSTU segmentation method:
the image is first converted to a grayscale image. The number of pixels in the image, of which the gray value A is smaller than the threshold T, is H0, and the number of pixels of which the gray value A is larger than the threshold T is H1; recording the average gray value of H0 as H0 and the average gray value of H1 as H1 in the image with the image size of X X Y and the threshold value of T;
the probability that the pixel gray value is less than T is:
r0＝h0/(X*Y)；
the probability that the pixel gray value is greater than T is:
r1＝h1/(X*Y)；
h0+h1＝X*Y；
r0+r1＝1；
the average gray is multiplied by the probability and then added:
e＝r0*h0+r1*h1；
the between-class variance is:
d＝r0(h0-e)^2+r1(h1-e)^2；
d＝r0*r1(h0-h1)^2
binary morphology of mathematical morphology
And (3) corrosion algorithm:
the method is simply understood as reducing the range of a target area, and the image boundary is contracted from the visual perception of the image; in practice, it is often used to eliminate noise or unwanted objects. The expression is as follows:
the expression is expressed as corroding the set A by the set B, namely, enabling B to move in A by a starting point and convolving with an overlapping area of A, and if the values at the position B and the position A are the same, outputting a result of 1, otherwise, outputting a result of 0. The flow chart of the binary morphological erosion algorithm is shown in fig. 12.
And (3) expansion algorithm:
simply comprehending that the range of a target area is enlarged, and the image boundary is expanded from the visual perception of the image; in practical application, the method is mainly used for filling the holes in the target area and eliminating small particle noise. The expression is as follows:
the above expression is expressed as using the set B to expand the set A, i.e. let B move in A with a starting point, and make convolution with the overlapping area of A, if the intersection of the values at the position B and the position A is not null, the output result is 1, otherwise, the output result is 0. The flow chart of the binary morphology inflation algorithm is shown in fig. 13.
Grayscale morphology of mathematical morphology
The image element is marked as A, the structural element is marked as B, and the area of the structural element covering the image is marked as C.
Etching of gray scale morphology:
simply understood as the operation of convolution, a small rectangle C formed by subtracting the structural element B from A is used, the minimum value in C is taken, and the minimum value is assigned to the origin corresponding to B. As shown in fig. 14, a gray scale morphological erosion algorithm flow chart.
Dilation algorithm for gray morphology:
simply understood as the operation of convolution, a small rectangle C formed by adding A and the structural element B is used, the maximum value in C is taken, and the original point corresponding to B is assigned. The gray scale morphological dilation algorithm flow chart shown in fig. 15.
Examples
(1) The infusion bag contains a foreign substance as shown in FIG. 16.
(2) Adopting an optical frequency characteristic image characteristic identification system; the bubble identification system comprises a camera 2 and a light source;
the light sources comprise a first light source 4, a second light source 6 and a third light source 8; the first light source 4 is mounted on a first mounting plate 10, the second light source 6 is mounted on a second mounting plate 11, and the third light source 8 is mounted on a third mounting plate 12;
the first light source 4 is controlled by a first industrial personal computer 5, the second light source 6 is controlled by a second industrial personal computer 7, and the third light source 8 is controlled by a third industrial personal computer 9;
the camera 2 is connected with a PC controller 3; the light source is connected with an industrial personal computer; the light source is used for emitting a light source with a specific optical frequency, a specific light-emitting duration, a specific position and a specific shape; the light emitted by the light source is emitted to the object 1 to be measured; the camera 2 is used for shooting the measured object 1; the PC controller 3 is used for collecting images and identifying the bubble space optical characteristics of the object 1 to be measured;
then the following steps are carried out:
s1, by modulating the radiant energy flux of the light source with single frequency, generating light with different energy amplitudes to illuminate the target object, generating images with different optical characteristics, and screening the characteristics of a series of images to achieve the purpose of identifying the target object; continuously modulating the radiant energy flux of a single light source to obtain continuously-changed analog quantity image data, and acquiring and identifying an optical characteristic image which is formed by a target object in a state of continuously modulating the radiant energy flux and continuously changes;
s2, adjusting the illumination time and the non-illumination time of the light source by controlling the light source switch, and carrying out time modulation on the light source with single frequency so as to obtain the images of the optical characteristics of the object with bright background and dark background at different illumination times;
s3 passing through three-dimensional coordinates X/Y/Z and pitch angle theta in adjustable three-dimensional space1And a rotation angle theta2The light source adjusting bracket adjusts the irradiation position of a single light source, obtains optical characteristics of the same target in different irradiation directions at different irradiation directions and distances, and further realizes the identification of the target object through the identification of the characteristics;
s4, using light source with multiple frequencies to irradiate,
according to the Grassmann color mixing law, the method is suitable for color light additive color mixing, namely light rays with different wavelengths are superposed, and light and shade change is carried out on colors with specified wavelengths through light superposition with different wavelengths, so that objects with specified colors are filtered and screened;
blue light (460nm) + yellow light (580nm) as white light
Red light (660nm) + cyan light (480nm) as white light
The imaging quality of different substances is different, and background noise is filtered by using a light source with combined frequency, so that the image contrast is enhanced, and the recognition rate is improved;
s5, based on the characteristics of the steps S1, S2, S3 and S4, performing combination matching on parameters such as radiation energy flux modulation, irradiation time modulation, distributed irradiation positions and the like of the multi-frequency light source to form structured light with a specific combination, and forming specific optical characteristics on the target object; as shown in fig. 16.
(3) The binary image obtained by the OSTU algorithm is shown in fig. 17.
(4) The image of the binary image after the erosion expansion processing is shown in fig. 18;
(5) after the bubble font feature detection is automatically performed in fig. 3, it is determined as a foreign object by an exclusion method, as shown in fig. 19.
Claims (3)
1. An image characteristic identification method based on optical frequency characteristics is characterized in that an optical frequency characteristic image characteristic identification system is adopted; the bubble identification system comprises a camera (2) and a light source;
the light source comprises a first light source (4), a second light source (6) and a third light source (8); the first light source (4) is mounted on a first mounting plate (10), the second light source (6) is mounted on a second mounting plate (11), and the third light source (8) is mounted on a third mounting plate (12);
the first light source (4) is controlled by a first industrial personal computer (5), the second light source (6) is controlled by a second industrial personal computer (7), and the third light source (8) is controlled by a third industrial personal computer (9);
the camera (2) is connected with a PC controller (3); the light source is connected with an industrial personal computer; the light source is used for emitting a light source with a specific optical frequency, a specific light-emitting duration, a specific position and a specific shape; the light emitted by the light source is emitted to the object to be measured (1); the camera (2) is used for shooting a measured object (1); the PC controller (3) is used for collecting images and identifying the bubble space optical characteristics of the measured object (1);
and comprises the steps of:
s1, by modulating the radiant energy flux of the light source with single frequency, generating light with different energy amplitudes to illuminate the target object, generating images with different optical characteristics, and screening the characteristics of a series of images to achieve the purpose of identifying the target object; continuously modulating the radiant energy flux of a single light source to obtain continuously-changed analog quantity image data, and acquiring and identifying an optical characteristic image which is formed by a target object in a state of continuously modulating the radiant energy flux and continuously changes;
s2, adjusting the illumination time and the non-illumination time of the light source by controlling the light source switch, and carrying out time modulation on the light source with single frequency so as to obtain the images of the optical characteristics of the object with bright background and dark background at different illumination times;
s3 passing through three-dimensional coordinates X/Y/Z and pitch angle theta in adjustable three-dimensional space1And a rotation angle theta2The light source adjusting bracket adjusts the irradiation position of a single light source, obtains optical characteristics of the same target in different irradiation directions at different irradiation directions and distances, and further realizes the identification of the target object through the identification of the characteristics;
s4, using light source with multiple frequencies to irradiate,
according to the Grassmann color mixing law, the method is suitable for color light additive color mixing, namely light rays with different wavelengths are superposed, and light and shade change is carried out on colors with specified wavelengths through light superposition with different wavelengths, so that objects with specified colors are filtered and screened;
blue light (460nm) + yellow light (580nm) as white light
Red light (660nm) + cyan light (480nm) as white light
The imaging quality of different substances is different, and background noise is filtered by using a light source with combined frequency, so that the image contrast is enhanced, and the recognition rate is improved;
s5, based on the characteristics of the steps S1, S2, S3 and S4, performing combination matching on parameters such as radiation energy flux modulation, irradiation time modulation, distributed irradiation positions and the like of the multi-frequency light source to form structured light with a specific combination, and forming specific optical characteristics on the target object;
and S6, automatically identifying the image features through a software algorithm.
2. The image feature recognition method based on optical frequency features as claimed in claim 1, wherein: in the step S6, the software algorithm adopts an OSTU segmentation method;
the method specifically comprises the following steps: firstly, converting an image into a gray scale image; the number of pixels in the image, of which the gray value A is smaller than the threshold T, is H0, and the number of pixels of which the gray value A is larger than the threshold T is H1; recording the average gray value of H0 as H0 and the average gray value of H1 as H1 in the image with the image size of X X Y and the threshold value of T;
the probability that the pixel gray value is less than T is:
r0＝h0/(X*Y)；
the probability that the pixel gray value is greater than T is:
r1＝h1/(X*Y)；
h0+h1＝X*Y；
r0+r1＝1；
the average gray is multiplied by the probability and then added:
e＝r0*h0+r1*h1；
the between-class variance is:
d＝r0(h0-e)^2+r1(h1-e)^2；
d＝r0*r1(h0-h1)^2。
3. the image feature recognition method based on optical frequency features as claimed in claim 1, wherein: the software algorithm adopts a corrosion algorithm, an expansion algorithm or an expansion algorithm of gray-scale morphology.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911148466.5A CN111027527A (en) | 2019-11-21 | 2019-11-21 | Image feature identification method based on optical frequency features |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911148466.5A CN111027527A (en) | 2019-11-21 | 2019-11-21 | Image feature identification method based on optical frequency features |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111027527A true CN111027527A (en) | 2020-04-17 |
Family
ID=70201800
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911148466.5A Pending CN111027527A (en) | 2019-11-21 | 2019-11-21 | Image feature identification method based on optical frequency features |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111027527A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114219841A (en) * | 2022-02-23 | 2022-03-22 | 武汉欧耐德润滑油有限公司 | Automatic identification method of lubricating oil tank parameters based on image processing |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101670636A (en) * | 2009-09-27 | 2010-03-17 | 上海大学 | Automatic bubble eliminating system for vacuum casting and method |
| CN104903914A (en) * | 2012-11-07 | 2015-09-09 | 生物梅里埃公司 | Bio-imaging method |
| CN110121021A (en) * | 2019-06-28 | 2019-08-13 | 四川极智朗润科技有限公司 | A kind of optical shutter system and its imaging method suitable for Rolling shutter camera |
| CN209542479U (en) * | 2019-06-18 | 2019-10-25 | 成都利尔药业有限公司 | A kind of two-way artificial candling apparatus of ampoule |
-
2019
- 2019-11-21 CN CN201911148466.5A patent/CN111027527A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101670636A (en) * | 2009-09-27 | 2010-03-17 | 上海大学 | Automatic bubble eliminating system for vacuum casting and method |
| CN104903914A (en) * | 2012-11-07 | 2015-09-09 | 生物梅里埃公司 | Bio-imaging method |
| CN209542479U (en) * | 2019-06-18 | 2019-10-25 | 成都利尔药业有限公司 | A kind of two-way artificial candling apparatus of ampoule |
| CN110121021A (en) * | 2019-06-28 | 2019-08-13 | 四川极智朗润科技有限公司 | A kind of optical shutter system and its imaging method suitable for Rolling shutter camera |
Non-Patent Citations (1)
| Title |
|---|
| 朱秀昌等: "《数字图像处理与图像信息》", 31 August 2016, 北京邮电大学出版社 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114219841A (en) * | 2022-02-23 | 2022-03-22 | 武汉欧耐德润滑油有限公司 | Automatic identification method of lubricating oil tank parameters based on image processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10102614B2 (en) | Fog removing device and image generating method | |
| CN103344563B (en) | A kind of self-adaptation toning light modulation machine vision light source pick-up unit and method | |
| CN102422322B (en) | Method and apparatus for reproducing the color appearance of an image from a source environment at a device in a target environment | |
| CN102867185B (en) | Method and system for identifying automobile tire number | |
| CN110216082B (en) | Method and system for dynamic identification of fluorescently labeled seeds | |
| CN108920992A (en) | A kind of positioning and recognition methods of the medical label bar code based on deep learning | |
| JP5914688B2 (en) | Color analysis for digital images | |
| Kwon et al. | Visible and near-infrared image acquisition and fusion for night surveillance | |
| CN109461186A (en) | Image processing method, device, computer readable storage medium and electronic equipment | |
| CN114450579B (en) | Image processing system, setting method and recording medium | |
| JP4870807B2 (en) | Edge detection method and image processing apparatus | |
| CN105354801A (en) | HSV color space based image enhancement method | |
| CN111027527A (en) | Image feature identification method based on optical frequency features | |
| ES2552405B1 (en) | System and method of detection of Anisakis parasites in fish fillets | |
| JP2021190802A (en) | Imaging device, inspection apparatus and imaging method | |
| CN110969153A (en) | Bubble identification method based on space optical characteristics | |
| CN109905677A (en) | A kind of coal mine monitoring image enhancement processing method | |
| CN118134886A (en) | Laser printed matter imaging defect detection method | |
| Coelho et al. | A machine vision system for automatic detection of parasites Edotea magellanica in shell-off cooked clam Mulinia edulis | |
| JP2022537196A (en) | surface recognition | |
| WO2023134304A1 (en) | Optical information collection apparatus and optical information collection method | |
| US10650205B2 (en) | Methods, systems, and apparatuses for scanning and decoding direct part marking indicia | |
| CN117037064B (en) | Detection method and system for illegal land occupation and tillage actions based on improved SSD | |
| JP2015501968A (en) | Post-processing to improve eigenimage generation | |
| CN117830331A (en) | Multi-Otsu threshold segmentation-based colony automatic counting method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200417 |