CN118392834B - Detection method for transparent flexible substrate of integrated circuit - Google Patents

Abstract

The present invention provides a detection method for an integrated circuit transparent flexible substrate. It relates to the technical field of substrate detection, and the method includes: performing performance detection on an integrated circuit transparent flexible substrate; obtaining a substrate image of an integrated circuit transparent flexible substrate that has passed the performance detection and performing size detection, and performing image preprocessing after the size detection is passed; performing image segmentation on the substrate image after image preprocessing, obtaining a substrate sub-image to extract the edge of the substrate line and fit the substrate line, determining the line width of the substrate line, and performing defect detection on the integrated circuit transparent flexible substrate; achieving efficient detection and high-quality detection of the integrated circuit transparent flexible substrate.

Description

A detection method for integrated circuit transparent flexible substrate

Technical Field

The present invention relates to the technical field of substrate detection, and in particular to a detection method for a transparent flexible substrate of an integrated circuit.

Background Art

With the development of electronic products, transparent flexible substrates are widely used in electronic products and develop rapidly due to their high transparency, light weight, good flexibility, ultra-thin characteristics and high-density integration. Integrated circuits achieve specific functions by designing and connecting multiple electronic components. However, with the continuous complexity of integrated circuits and the demand for lighter, thinner and more transparent electronic products, the manufacturing requirements of integrated circuit transparent flexible substrates are increasing, and the detection of integrated circuit transparent flexible substrates is becoming more difficult and inefficient.

Therefore, the present invention provides a detection method for a transparent flexible substrate of an integrated circuit.

Summary of the invention

The present invention provides a detection method for an integrated circuit transparent flexible substrate, which is used for performing performance detection on the integrated circuit transparent flexible substrate; obtaining a substrate image of the integrated circuit transparent flexible substrate that has passed the performance detection and performing size detection, and performing image preprocessing when the size detection is qualified; performing image segmentation on the substrate image after the image preprocessing, obtaining a substrate sub-image to extract the edge of the substrate line and fitting the substrate line, determining the line width of the substrate line, and performing defect detection on the integrated circuit transparent flexible substrate; thus, efficient detection and high-quality detection of the integrated circuit transparent flexible substrate are achieved.

The present invention provides a detection method for an integrated circuit transparent flexible substrate, comprising:

Step 1: Performing performance testing on the integrated circuit transparent flexible substrate, wherein the performance testing includes: integrated circuit performance testing, transparency performance testing and flexibility performance testing;

Step 2: Obtain a substrate image of the integrated circuit transparent flexible substrate that has passed the performance test and perform size detection, and perform image preprocessing when the size detection is qualified;

Step 3: Perform image segmentation on the substrate image after image preprocessing, obtain the substrate sub-image, extract the substrate line edge and fit the substrate line, determine the substrate line width, and perform defect detection on the integrated circuit transparent flexible substrate.

According to a method for detecting a transparent flexible substrate for an integrated circuit provided by the present invention, a performance detection is performed on the transparent flexible substrate for the integrated circuit, comprising:

Performing transparency performance testing on the integrated circuit transparent flexible substrate according to the pre-deployed transparency testing equipment, and when monitoring that the transparency of the integrated circuit transparent flexible substrate in the visible light range is not less than a preset transparency, determining that the transparency performance test of the integrated circuit transparent flexible substrate is qualified;

Fixing one end of the integrated circuit transparent flexible substrate, gradually bending the other end of the integrated circuit transparent flexible substrate according to a preset flexibility detection angle, and performing a flexibility performance test on the integrated circuit transparent flexible substrate;

An integrated circuit performance test is performed on the integrated circuit transparent flexible substrate that has passed the transparency performance test and the flexibility performance test. When it is monitored that there is no short circuit or open circuit in the circuit of the integrated circuit transparent flexible substrate, it is determined that the integrated circuit performance test of the integrated circuit transparent flexible substrate has passed.

According to a method for detecting a transparent flexible substrate for an integrated circuit provided by the present invention, a substrate image of a transparent flexible substrate for an integrated circuit that has passed a performance test is obtained and a size test is performed, comprising:

Pre-deploy an inspection camera for the integrated circuit transparent flexible substrate and determine an image acquisition area for the inspection camera;

Calibrate and cover the image acquisition area of the detection camera according to the preset precision calibration plate, select a corner of the preset precision calibration plate as the coordinate origin, and establish the image acquisition area coordinate system;

When it is monitored that the integrated circuit transparent flexible substrate that has passed the performance test completely enters the image acquisition area of the detection camera, an image of the substrate is acquired;

Acquire pixel coordinates of each pixel point of the integrated circuit transparent flexible substrate in the substrate image according to the image acquisition area coordinate system, determine edge pixel points of the integrated circuit transparent flexible substrate, and then determine first length data and first width data of the integrated circuit transparent flexible substrate;

Detecting second length data, second width data, and second thickness data of the integrated circuit transparent flexible substrate using a preset optical measuring device, and if the second length data is consistent with the first length data and the second width data is consistent with the first width data, recording the second length data, the second width data, and the second thickness data as the size data of the integrated circuit transparent flexible substrate;

Otherwise, the second thickness data is corrected according to the data error, and the first length data, the first width data and the second thickness data after the data correction are recorded as the size data of the transparent flexible substrate of the integrated circuit;

When the size data of the integrated circuit transparent flexible substrate meets the preset size specification, it is determined that the size detection is qualified.

According to a detection method for an integrated circuit transparent flexible substrate provided by the present invention, image preprocessing is performed after the size detection is qualified, comprising:

Perform LAB color space conversion on the substrate image that has passed the size inspection to obtain the LAB color space of the substrate image;

Performing image filtering processing on a and b components of the LAB color space of the substrate image, and determining the total number of iterations of the image filtering processing according to the size data of the integrated circuit transparent flexible substrate;

When it is monitored that the number of iterations is not less than the total number of iterations of the image filtering process, the image filtering process is determined to be completed, and the substrate image after the last iteration is completed is obtained, which are recorded as the a component filtering image and the b component filtering image respectively;

Otherwise, repeat the iteration to perform image filtering;

Performing image enhancement processing on the L component of the LAB color space of the substrate image to obtain an L component enhanced image;

The a component filtered image, the b component filtered image and the L component enhanced image are merged and converted into RGB color space to determine a substrate image after preprocessing.

According to a detection method for an integrated circuit transparent flexible substrate provided by the present invention, an image enhancement process is performed on the L component of the LAB color space of the substrate image to obtain an L component enhanced image, including:

Obtain the pixel value of each pixel point of the L component of the LAB color space of the substrate image. When the pixel value is not lower than the preset pixel threshold, it is marked as a background pixel to extract the background low-illuminance area;

Decomposing the illumination image and the reflection image of the background low illumination area, performing grayscale enhancement processing on the illumination image, and performing filtering processing on the reflection image;

The illumination image after grayscale enhancement processing and the reflection image after filtering processing are combined to obtain an L component enhanced image.

According to a detection method for an integrated circuit transparent flexible substrate provided by the present invention, image segmentation is performed on a substrate image after image preprocessing, comprising:

Extracting pixel grayscale features of the substrate image after image preprocessing, and determining a pixel grayscale segmentation threshold to perform threshold segmentation on the substrate image after image preprocessing;

Acquire edge pixels in the substrate image after the image preprocessing, and if the coordinate difference of the edge pixels is not higher than the preset coordinate difference, connect the corresponding edge pixels to achieve edge segmentation of the substrate image after the image preprocessing;

Clustering and merging the pixel points of the substrate image after the image preprocessing according to the substrate features, and clustering and segmenting the substrate image after the image preprocessing according to the clustering and merging result, wherein the substrate features include: color features, grayscale features and spatial features;

The threshold segmentation results, edge segmentation results and cluster segmentation results are comprehensively processed, and the substrate image after image preprocessing is segmented according to the comprehensive processing results to obtain a substrate sub-image.

A detection method for an integrated circuit transparent flexible substrate provided by the present invention further includes:

Constructing an image segmentation performance evaluation index set, and performing image segmentation performance evaluation and quantitative processing on the comprehensive processing result, wherein the image segmentation performance evaluation index set includes: a pixel segmentation accuracy evaluation index, a pixel error segmentation rate evaluation index, a substrate sub-image true similarity rate evaluation index, and a substrate sub-image information rate evaluation index;

When the image segmentation performance evaluation quantization value is not lower than the preset evaluation value, the image segmentation is determined to be accurate.

According to a detection method for an integrated circuit transparent flexible substrate provided by the present invention, the line width of the substrate is determined, and defects of the integrated circuit transparent flexible substrate are detected, including:

Extracting the edge of the substrate circuit on the substrate sub-image to obtain the substrate circuit;

When it is monitored that the line length of the substrate line is not greater than the line length threshold, the corresponding substrate line is determined to be a substrate line with no impact defects, and the line is removed;

Constructing a substrate circuit set according to the substrate circuits after circuit elimination, fitting and classifying the substrate circuits in the substrate circuit set respectively, randomly selecting a pixel point of the substrate circuit, obtaining the change difference of the pixel slope on the left side and the pixel slope on the right side of the pixel point, and determining the pixel slope of the pixel point;

Determine the pixel slopes of each pixel point of the substrate line according to the pixel slope of the one pixel point, and when the pixel slopes of each pixel point of the substrate line are not higher than a preset slope threshold, determine that the corresponding substrate line is a substrate straight line and perform fitting processing to obtain a fitting straight line;

Otherwise, locate the pixel point of the substrate line whose pixel slope is higher than the preset slope threshold, and determine that the corresponding substrate line is a substrate curved line;

The substrate curved road is divided according to the central pixel point of the substrate curved road, and the substrate straight road is repeatedly determined and fitted to obtain a fitting straight road;

When it is detected that the substrate straight line is repeatedly determined and there is no substrate curved line, the division is terminated;

When the difference in the slopes of the fitted straight lines of the fitted straight lines is not higher than a preset difference threshold, the corresponding fitted straight lines are determined to be adjacent fitted straight lines;

Determine an average substrate line width of the integrated circuit transparent flexible substrate according to adjacent fitted straight lines;

; Wherein, F represents the average substrate line width of the integrated circuit transparent flexible substrate; Indicates one of the adjacent fitted straight lines; Indicates another fitted straight line in the adjacent fitted straight lines; Indicates the j1th pixel point in the 1 fitting straight line The first-order pixel slope in the x-axis direction; Indicates the j1th pixel point in the 1 fitting straight line The first-order pixel slope in the y-axis direction; Indicates the j1th pixel point in the 1 fitting straight line The second-order pixel slope in the x-direction; Indicates the j1th pixel point in the 1 fitting straight line The second-order pixel slope in the y-axis direction; n1 represents the number of adjacent fitting straight lines of the integrated circuit transparent flexible substrate; mi1 represents the number of pixel points of the one fitting straight line of the i1th adjacent fitting straight line in the integrated circuit transparent flexible substrate;

Performing defect detection on the integrated circuit transparent flexible substrate, if the average substrate line width of the integrated circuit transparent flexible substrate is consistent with the standard substrate line width or the line width error is not greater than the line width error threshold, then determining that the integrated circuit transparent flexible substrate line width detection is qualified;

Otherwise, the corresponding pixel point whose line width error is greater than the line width error threshold is a defective pixel. If the defective pixel ratio is not higher than the preset defect ratio, the line width detection of the integrated circuit transparent flexible substrate is determined to be qualified.

Compared with the prior art, the beneficial effects of this application are as follows:

The integrated circuit transparent flexible substrate is subjected to performance testing; the substrate image of the integrated circuit transparent flexible substrate that has passed the performance testing is obtained and the size testing is performed, and the image preprocessing is performed after the size testing is passed; the substrate image after the image preprocessing is performed with image segmentation, and the substrate sub-image is obtained to extract the substrate line edge and fit the substrate line to determine the substrate line width, and the integrated circuit transparent flexible substrate is subjected to defect detection; the efficient and high-quality testing of the integrated circuit transparent flexible substrate is achieved.

Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

FIG. 1 is a schematic flow chart of a method for detecting a transparent flexible substrate for an integrated circuit provided in an embodiment of the present invention.

DETAILED DESCRIPTION

In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

Embodiment 1:

As shown in FIG1 , an embodiment of the present invention provides a method for detecting a transparent flexible substrate for an integrated circuit, and the method mainly includes the following steps:

Step 1: Performing performance testing on the integrated circuit transparent flexible substrate, wherein the performance testing includes: integrated circuit performance testing, transparency performance testing and flexibility performance testing;

Step 2: Obtain a substrate image of the integrated circuit transparent flexible substrate that has passed the performance test and perform size detection, and perform image preprocessing when the size detection is qualified;

Step 3: Perform image segmentation on the substrate image after image preprocessing, obtain the substrate sub-image, extract the substrate line edge and fit the substrate line, determine the substrate line width, and perform defect detection on the integrated circuit transparent flexible substrate.

In this embodiment, the integrated circuit flexible transparent substrate refers to a base material used for electronic devices, which has high flexibility, high transparency and high-density integration.

In this embodiment, the integrated circuit performance detection, for example, detects whether the integrated circuit flexible transparent substrate has a short circuit or a break circuit.

In this embodiment, the transparency performance is detected, for example, the transparency of the integrated circuit flexible transparent substrate in the visible light range is detected.

In this embodiment, the flexibility performance is tested, for example, the degree of bendability of the flexible transparent substrate of the integrated circuit when one end is fixed.

In this embodiment, the substrate image of the integrated circuit transparent flexible substrate that has passed the performance test refers to an image obtained by photographing the integrated circuit transparent flexible substrate with a test camera, including a complete integrated circuit transparent flexible substrate.

In this embodiment, the size detection, for example, detects that the transparent flexible substrate of the integrated circuit is within a preset size range.

In this embodiment, the size detection includes substrate length size detection, substrate width size detection and substrate thickness size detection.

In this embodiment, image segmentation is performed on the substrate image after image preprocessing, so as to facilitate subsequent line edge extraction.

In this embodiment, the substrate sub-image refers to an image obtained by performing image segmentation on the substrate image after image preprocessing.

In this embodiment, fitting the substrate circuit refers to fitting the substrate circuit into a substrate straight line or fitting the substrate curved line.

In this embodiment, the substrate line width refers to the line distance of the integrated circuit transparent flexible substrate determined according to adjacent fitting straight lines.

In this embodiment, defect detection is performed on the integrated circuit transparent flexible substrate, for example, defective pixels of the integrated circuit transparent flexible substrate are detected.

The beneficial effects of the above technical scheme are: performing performance testing on the integrated circuit transparent flexible substrate; obtaining a substrate image of the integrated circuit transparent flexible substrate that has passed the performance test and performing size testing, and performing image preprocessing when the size test is qualified; performing image segmentation on the substrate image after image preprocessing, obtaining a substrate sub-image to extract the substrate line edge and fit the substrate line, determining the substrate line width, and performing defect detection on the integrated circuit transparent flexible substrate; achieving efficient and high-quality detection of the integrated circuit transparent flexible substrate.

Embodiment 2:

The embodiment of the present invention provides a method for detecting a transparent flexible substrate for an integrated circuit, and performs performance detection on the transparent flexible substrate for an integrated circuit, comprising:

Performing transparency performance testing on the integrated circuit transparent flexible substrate according to the pre-deployed transparency testing equipment, and when monitoring that the transparency of the integrated circuit transparent flexible substrate in the visible light range is not less than a preset transparency, determining that the transparency performance test of the integrated circuit transparent flexible substrate is qualified;

Fixing one end of the integrated circuit transparent flexible substrate, gradually bending the other end of the integrated circuit transparent flexible substrate according to a preset flexibility detection angle, and performing a flexibility performance test on the integrated circuit transparent flexible substrate;

An integrated circuit performance test is performed on the integrated circuit transparent flexible substrate that has passed the transparency performance test and the flexibility performance test. When it is monitored that there is no short circuit or open circuit in the circuit of the integrated circuit transparent flexible substrate, it is determined that the integrated circuit transparent flexible substrate has passed the integrated circuit performance test.

In this embodiment, one end of the integrated circuit transparent flexible substrate is fixed, and usually the short end of the integrated circuit transparent flexible substrate is selected for fixing.

In this embodiment, the other end of the integrated circuit transparent flexible substrate is gradually bent according to a preset flexibility detection angle, and stays at the preset flexibility detection angle for 3 seconds to ensure the accuracy of the flexibility performance detection.

In this embodiment, whether the circuit of the transparent flexible substrate of the integrated circuit is short-circuited or open-circuited can be determined by detecting the resistance and the capacitance.

The beneficial effect of the above technical solution is: the performance of the integrated circuit transparent flexible substrate is tested, which effectively ensures that the transparency performance, flexibility performance and integrated circuit performance of the integrated circuit transparent flexible substrate meet the preset substrate production requirements.

Embodiment 3:

The embodiment of the present invention provides a method for detecting a transparent flexible substrate for an integrated circuit, which obtains a substrate image of a transparent flexible substrate for an integrated circuit that has passed a performance test and performs a size test, including:

Pre-deploy an inspection camera for the integrated circuit transparent flexible substrate and determine an image acquisition area for the inspection camera;

Calibrate and cover the image acquisition area of the detection camera according to the preset precision calibration plate, select a corner of the preset precision calibration plate as the coordinate origin, and establish the image acquisition area coordinate system;

When it is monitored that the integrated circuit transparent flexible substrate that has passed the performance test completely enters the image acquisition area of the detection camera, an image of the substrate is acquired;

Acquire pixel coordinates of each pixel point of the integrated circuit transparent flexible substrate in the substrate image according to the image acquisition area coordinate system, determine edge pixel points of the integrated circuit transparent flexible substrate, and then determine first length data and first width data of the integrated circuit transparent flexible substrate;

Detecting second length data, second width data, and second thickness data of the integrated circuit transparent flexible substrate using a preset optical measuring device, and if the second length data is consistent with the first length data and the second width data is consistent with the first width data, recording the second length data, the second width data, and the second thickness data as the size data of the integrated circuit transparent flexible substrate;

Otherwise, the second thickness data is corrected according to the data error, and the first length data, the first width data and the second thickness data after the data correction are recorded as the size data of the transparent flexible substrate of the integrated circuit;

When the size data of the integrated circuit transparent flexible substrate meets the preset size specification, it is determined that the size detection is qualified.

In this embodiment, the inspection camera of the integrated circuit transparent flexible substrate is generally pre-deployed directly above the integrated circuit transparent flexible substrate.

In this embodiment, the preset accuracy calibration plate is used to determine the coordinates of the image acquisition area of the detection camera. The higher the detection accuracy of the integrated circuit transparent flexible substrate, the higher the preset accuracy of the calibration plate.

In this embodiment, the first length data and the first width data refer to length data and width data determined according to edge pixels of the transparent flexible substrate of the integrated circuit.

In this embodiment, the second length data, the second width data and the second thickness data refer to data determined according to a preset optical measuring device.

In this embodiment, the detection camera cannot obtain thickness data of the transparent flexible substrate of the integrated circuit.

In this embodiment, the dimension data acquired by the preset optical measuring device may have data errors, and the data correction is performed by detecting the dimension data acquired by the camera.

In this embodiment, the dimension specifications are preset, for example, the length dimension specification is limited to a1-a2.

The beneficial effects of the above technical solution are: obtaining a substrate image of an integrated circuit transparent flexible substrate that has passed the performance test, which is convenient for subsequent defect detection of the integrated circuit transparent flexible substrate; and performing size detection to further ensure that the integrated circuit flexible transparent substrate meets the preset substrate production requirements.

Embodiment 4:

The embodiment of the present invention provides a method for detecting a transparent flexible substrate for an integrated circuit, and when the size detection is qualified, image preprocessing is performed, comprising:

Perform LAB color space conversion on the substrate image that has passed the size inspection to obtain the LAB color space of the substrate image;

Performing image filtering processing on a and b components of the LAB color space of the substrate image, and determining the total number of iterations of the image filtering processing according to the size data of the integrated circuit transparent flexible substrate;

When it is monitored that the number of iterations is not less than the total number of iterations of the image filtering process, the image filtering process is determined to be completed, and the substrate image after the last iteration is completed is obtained, which are recorded as the a component filtering image and the b component filtering image respectively;

Otherwise, repeat the iteration to perform image filtering;

Performing image enhancement processing on the L component of the LAB color space of the substrate image to obtain an L component enhanced image;

The a component filtered image, the b component filtered image and the L component enhanced image are merged and converted into RGB color space to determine a substrate image after preprocessing.

In this embodiment, the a component of the substrate image LAB color space represents the component from green to red; the b component represents the component from blue to yellow; and the L component represents brightness.

In this embodiment, the image filtering process refers to processing according to an adaptive filtering algorithm.

In this embodiment, the a-component filtered image, the b-component filtered image and the L-component enhanced image are merged to effectively improve the image quality of the substrate image.

The beneficial effect of the above technical solution is that: when the size detection is qualified, image preprocessing is performed to facilitate the subsequent image segmentation of the substrate image, which effectively improves the subsequent image segmentation accuracy and image segmentation efficiency.

Embodiment 5:

The embodiment of the present invention provides a detection method for an integrated circuit transparent flexible substrate, which performs image enhancement processing on the L component of the LAB color space of the substrate image to obtain an L component enhanced image, including:

Obtain the pixel value of each pixel point of the L component of the LAB color space of the substrate image. When the pixel value is not lower than the preset pixel threshold, it is marked as a background pixel to extract the background low-illuminance area;

Decomposing the illumination image and the reflection image of the background low illumination area, performing grayscale enhancement processing on the illumination image, and performing filtering processing on the reflection image;

The illumination image after grayscale enhancement processing and the reflection image after filtering processing are combined to obtain an L component enhanced image.

In this embodiment, background low-illuminance area extraction is performed to effectively avoid subsequent image enhancement processing that may lead to loss of image details.

In this embodiment, multiple iterative decompositions are required to decompose the illumination image and the reflection image of the background low illumination area.

In this embodiment, performing illumination image decomposition and reflection image decomposition on the background low illumination area can effectively avoid enhancing image noise during subsequent image enhancement processing.

In this embodiment, a decomposition algorithm is selected to decompose the illumination image and the reflection image of the background low illumination area, for example, an SSR decomposition algorithm and an MSR decomposition algorithm.

In this embodiment, the illumination image is subjected to grayscale enhancement processing, and the reflection image is subjected to filtering processing, so as to improve the image quality.

The beneficial effect of the above technical solution is: performing image enhancement processing on the L component of the LAB color space of the substrate image to obtain an L component enhanced image, effectively improving the image contrast, and facilitating subsequent defect detection of the integrated circuit transparent flexible substrate.

Embodiment 6:

The embodiment of the present invention provides a detection method for a transparent flexible substrate of an integrated circuit, which performs image segmentation on a substrate image after image preprocessing, comprising:

Extracting pixel grayscale features of the substrate image after image preprocessing, and determining a pixel grayscale segmentation threshold to perform threshold segmentation on the substrate image after image preprocessing;

Acquire edge pixels in the substrate image after the image preprocessing, and if the coordinate difference of the edge pixels is not higher than the preset coordinate difference, connect the corresponding edge pixels to achieve edge segmentation of the substrate image after the image preprocessing;

Clustering and merging the pixel points of the substrate image after the image preprocessing according to the substrate features, and clustering and segmenting the substrate image after the image preprocessing according to the clustering and merging result, wherein the substrate features include: color features, grayscale features and spatial features;

The threshold segmentation results, edge segmentation results and cluster segmentation results are comprehensively processed, and the substrate image after image preprocessing is segmented according to the comprehensive processing results to obtain a substrate sub-image.

In this embodiment, the pixel grayscale segmentation threshold may be determined according to the average grayscale threshold of the substrate image after image preprocessing.

In this embodiment, comprehensive processing of the threshold segmentation results, edge segmentation results and cluster segmentation results refers to selective segmentation based on the substrate image. For example, if the pixel difference between the integrated circuit transparent flexible substrate pixel and the background pixel is less than a1, the threshold segmentation weight is reduced and the edge segmentation weight is increased to process the substrate image.

The beneficial effect of the above technical solution is: by performing image segmentation on the substrate image after image preprocessing, a substrate sub-image is obtained, which is conducive to the subsequent substrate line edge extraction and fitting processing of the substrate line, thereby improving the subsequent detection efficiency and detection quality of the integrated circuit transparent flexible substrate.

Embodiment 7:

An embodiment of the present invention provides a detection method for an integrated circuit transparent flexible substrate, further comprising:

Constructing an image segmentation performance evaluation index set, and performing image segmentation performance evaluation and quantitative processing on the comprehensive processing result, wherein the image segmentation performance evaluation index set includes: a pixel segmentation accuracy evaluation index, a pixel error segmentation rate evaluation index, a substrate sub-image true similarity rate evaluation index, and a substrate sub-image information rate evaluation index;

When the image segmentation performance evaluation quantization value is not lower than the preset evaluation value, the image segmentation is determined to be accurate.

In this embodiment, the pixel segmentation accuracy refers to the accuracy of correctly segmenting the pixels of the transparent flexible substrate of the integrated circuit and the background pixels.

In this embodiment, the weights of the evaluation indicators in the image segmentation performance evaluation indicator set are different and can be dynamically adjusted according to the production requirements of the integrated circuit transparent flexible substrate.

The beneficial effect of the above technical solution is: by constructing an image segmentation performance evaluation index set, the image segmentation performance evaluation and quantitative processing are performed on the comprehensive processing results, which facilitates accurate evaluation of image segmentation quality and thus improves image segmentation quality.

Embodiment 8:

The embodiment of the present invention provides a detection method for an integrated circuit transparent flexible substrate, which determines the line width of the substrate line and performs defect detection on the integrated circuit transparent flexible substrate, including:

Extracting the edge of the substrate circuit on the substrate sub-image to obtain the substrate circuit;

When it is monitored that the line length of the substrate line is not greater than the line length threshold, the corresponding substrate line is determined to be a substrate line with no impact defects, and the line is removed;

Constructing a substrate circuit set according to the substrate circuits after circuit elimination, fitting and classifying the substrate circuits in the substrate circuit set respectively, randomly selecting a pixel point of the substrate circuit, obtaining the change difference of the pixel slope on the left side and the pixel slope on the right side of the pixel point, and determining the pixel slope of the pixel point;

Determine the pixel slopes of each pixel point of the substrate line according to the pixel slope of the one pixel point, and when the pixel slopes of each pixel point of the substrate line are not higher than a preset slope threshold, determine that the corresponding substrate line is a substrate straight line and perform fitting processing to obtain a fitting straight line;

Otherwise, locate the pixel point of the substrate line whose pixel slope is higher than the preset slope threshold, and determine that the corresponding substrate line is a substrate curved line;

The substrate curved road is divided according to the central pixel point of the substrate curved road, and the substrate straight road is repeatedly determined and fitted to obtain a fitting straight road;

When it is detected that the substrate straight line is repeatedly determined and there is no substrate curved line, the division is terminated;

When the difference in the slopes of the fitted straight lines of the fitted straight lines is not higher than a preset difference threshold, the corresponding fitted straight lines are determined to be adjacent fitted straight lines;

Determine an average substrate line width of the integrated circuit transparent flexible substrate according to adjacent fitted straight lines;

; Wherein, F represents the average substrate line width of the integrated circuit transparent flexible substrate; Indicates one of the adjacent fitted straight lines; Indicates another fitted straight line in the adjacent fitted straight lines; Indicates the j1th pixel point in the 1 fitting straight line The first-order pixel slope in the x-axis direction; Indicates the j1th pixel point in the 1 fitting straight line The first-order pixel slope in the y-axis direction; Indicates the j1th pixel point in the 1 fitting straight line The second-order pixel slope in the x-direction; Indicates the j1th pixel point in the 1 fitting straight line The second-order pixel slope in the y-axis direction; n1 represents the number of adjacent fitting straight lines of the integrated circuit transparent flexible substrate; mi1 represents the number of pixel points of the 1 fitting straight line of the i1th adjacent fitting straight line in the integrated circuit transparent flexible substrate;

Performing defect detection on the integrated circuit transparent flexible substrate, if the average substrate line width of the integrated circuit transparent flexible substrate is consistent with the standard substrate line width or the line width error is not greater than the line width error threshold, then determining that the integrated circuit transparent flexible substrate line width detection is qualified;

Otherwise, the corresponding pixel point whose line width error is greater than the line width error threshold is marked as a defective pixel. If the defective pixel ratio is not higher than the preset defect ratio, the line width detection of the integrated circuit transparent flexible substrate is determined to be qualified.

In this embodiment, fitting and classifying the substrate circuits in the substrate circuit set refers to fitting and classifying them into substrate straight lines and substrate curved lines.

In this embodiment, the substrate circuits that are determined to have no impact on the defective substrate circuits are eliminated, which is beneficial to improving the subsequent detection efficiency.

In this embodiment, the difference between the change of the pixel slope on the left side and the pixel slope on the right side of the pixel point is obtained to accurately determine the pixel slope of the pixel point.

In this embodiment, pixel points of the substrate lines whose pixel slopes are higher than a preset slope threshold are located. For example, substrate lines a1-a2-a3-a4, a2 and a3 are pixel points of the substrate lines whose pixel slopes are higher than a preset slope threshold, then substrate lines a1-a2 and a3-a4 are substrate straight lines; and substrate line a2-a3 is a substrate curved line.

In this embodiment, the substrate straight path determination is repeated to accurately detect and position the substrate straight path.

The beneficial effect of the above technical solution is: by extracting the substrate line edge of the substrate sub-image and fitting the substrate line, the substrate line width is determined, and defects are detected on the integrated circuit transparent flexible substrate, thereby improving the detection efficiency and detection quality.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (5)

1. A method for inspecting a transparent flexible substrate for an integrated circuit, comprising:

Step 1: performing performance detection on the integrated circuit transparent flexible substrate, wherein the performance detection comprises: integrated circuit performance detection, transparency performance detection, and flexibility performance detection;

step 2: acquiring a substrate image of the integrated circuit transparent flexible substrate with qualified performance detection, performing size detection, and performing image preprocessing after the size detection is qualified;

Step 3: image segmentation is carried out on the substrate image subjected to image pretreatment, substrate line edge extraction is carried out on the obtained substrate sub-image, a substrate line is subjected to fitting treatment, the line width of the substrate line is determined, and defect detection is carried out on the transparent flexible substrate of the integrated circuit, and the method specifically comprises the following steps:

Extracting pixel gray scale characteristics of a substrate image subjected to image pretreatment, and determining a pixel gray scale segmentation threshold value to carry out threshold segmentation on the substrate image subjected to image pretreatment;

Acquiring edge pixels in the substrate image subjected to the image preprocessing, and if the coordinate difference value of the edge pixels is not higher than a preset coordinate difference value, connecting the corresponding edge pixels to realize edge segmentation of the substrate image subjected to the image preprocessing;

Clustering and merging pixel points of the substrate image subjected to the image preprocessing according to substrate characteristics, and clustering and segmenting the substrate image subjected to the image preprocessing according to a clustering and merging result, wherein the substrate characteristics comprise: color features, gray features, and spatial features;

Comprehensively processing the threshold segmentation result, the edge segmentation result and the clustering segmentation result, and carrying out image segmentation on the substrate image subjected to the image preprocessing according to the comprehensive processing result to obtain a substrate sub-image;

Constructing an image segmentation performance evaluation index set, and performing image segmentation performance evaluation quantization processing on the comprehensive processing result, wherein the image segmentation performance evaluation index set comprises: the method comprises the steps of evaluating pixel point segmentation accuracy, evaluating pixel point error segmentation rate, evaluating real similarity of a substrate sub-image and evaluating information rate of the substrate sub-image;

when the image segmentation performance evaluation quantization value is not lower than a preset evaluation value, judging that the image segmentation is accurate;

Extracting the edges of the substrate lines of the substrate sub-images, obtaining the substrate lines, removing the substrate lines without influencing defects, and constructing a substrate line set;

Randomly selecting a pixel point of the substrate circuit, obtaining the change difference of the pixel slopes at two sides to determine the pixel slope and further determining the pixel slope of each pixel point of the substrate circuit;

when the pixel slope of each pixel point of the substrate circuit is not higher than a preset slope threshold, judging that the corresponding substrate circuit is a substrate straight circuit and performing fitting treatment to obtain a fitting straight circuit;

Otherwise, judging the corresponding substrate circuit as a substrate curved circuit and repeatedly judging the substrate straight circuit according to the central pixel point;

When the slope difference value of the fitting straight line is not higher than a preset difference value threshold value, judging that the corresponding fitting straight line is an adjacent fitting straight line, and determining the line width of the average substrate line of the transparent flexible substrate of the integrated circuit;

; wherein F represents the average substrate line width of the transparent flexible substrate of the integrated circuit; representing 1 fitting straight line in adjacent fitting straight lines; representing another 1 fitting straight line in the adjacent fitting straight lines; represents the j1 st pixel point in the 1 fitting straight line A first order pixel slope in the x-axis direction; represents the j1 st pixel point in the 1 fitting straight line A first order pixel slope in the y-axis direction; represents the j1 st pixel point in the 1 fitting straight line Second order pixel slope in the x-axis direction; represents the j1 st pixel point in the 1 fitting straight line Second order pixel slope in the y-axis direction; n1 represents the number of adjacent fitting straight lines of the transparent flexible substrate of the integrated circuit; mi1 represents the number of pixel points of the 1 st fitting straight line of the i1 st adjacent fitting straight lines in the transparent flexible substrate of the integrated circuit;

performing defect detection on the transparent flexible substrate of the integrated circuit, and if the line width of the line of the average substrate of the transparent flexible substrate of the integrated circuit is consistent with the line width of the line of the standard substrate or the line width error is not greater than the line width error threshold value, judging that the line width of the line of the transparent flexible substrate of the integrated circuit is qualified;

Otherwise, marking the corresponding pixel point with the line width error larger than the line width error threshold as a defective pixel, and judging that the line width detection of the transparent flexible substrate circuit of the integrated circuit is qualified if the defective pixel is not higher than the preset defective proportion.

2. The inspection method for transparent flexible substrates for integrated circuits according to claim 1, wherein performing performance inspection on the transparent flexible substrates for integrated circuits comprises:

Detecting transparency performance of the transparent flexible substrate of the integrated circuit according to pre-deployment transparency detection equipment, and judging that the transparency performance of the transparent flexible substrate of the integrated circuit is qualified when the transparency of the transparent flexible substrate of the integrated circuit in a visible light range is not lower than a preset transparency;

Fixing one end of the transparent flexible substrate of the integrated circuit, gradually bending the other end of the transparent flexible substrate of the integrated circuit according to a preset flexible detection angle, and detecting the flexible performance of the transparent flexible substrate of the integrated circuit;

And performing integrated circuit performance detection on the integrated circuit transparent flexible substrate with qualified transparency performance and finished flexible performance detection, and judging that the integrated circuit transparent flexible substrate has qualified integrated circuit performance detection when the circuit of the integrated circuit transparent flexible substrate is monitored to have no short circuit or open circuit.

3. The inspection method for transparent flexible substrates for integrated circuits according to claim 1, wherein acquiring substrate images of transparent flexible substrates for integrated circuits with acceptable performance inspection and performing dimension inspection comprises:

Pre-deploying a detection camera of the transparent flexible substrate of the integrated circuit, and determining an image acquisition area of the detection camera;

According to a preset precision calibration plate, calibrating and covering an image acquisition area of the detection camera, selecting one corner of the preset precision calibration plate as a coordinate origin, and establishing an image acquisition area coordinate system;

when the transparent flexible substrate of the integrated circuit with qualified performance detection is monitored to completely enter an image acquisition area of the detection camera, a substrate image is acquired;

Acquiring pixel coordinates of each pixel point of the transparent flexible substrate of the integrated circuit in the substrate image according to the image acquisition area coordinate system, determining edge pixel points of the transparent flexible substrate of the integrated circuit, and further determining first length data and first width data of the transparent flexible substrate of the integrated circuit;

Detecting second length data, second width data and second thickness data of the transparent flexible substrate of the integrated circuit according to preset optical measurement equipment, and if the second length data are consistent with the first length data and the second width data and the first width data, recording the second length data, the second width data and the second thickness data as size data of the transparent flexible substrate of the integrated circuit;

otherwise, carrying out data correction on the second thickness data according to the data error, and recording the first length data, the first width data and the second thickness data after the data correction as the size data of the transparent flexible substrate of the integrated circuit;

and when the size data of the transparent flexible substrate of the integrated circuit meets the preset size specification, judging that the size detection is qualified.

4. The inspection method for transparent flexible substrates for integrated circuits according to claim 1, wherein the image preprocessing is performed after the size inspection is qualified, comprising:

performing LAB color space conversion on the substrate image with qualified size detection to obtain a substrate image LAB color space;

carrying out image filtering processing on the component a and component b of the color space of the substrate image LAB, and determining the total number of image filtering processing iterations according to the size data of the transparent flexible substrate of the integrated circuit;

When the iteration times are not lower than the total iteration times of the image filtering processing, judging that the image filtering processing is completed, and acquiring a substrate image which is completed by the last iteration processing, wherein the substrate image is respectively marked as an a-component filtering image and a b-component filtering image;

otherwise, repeating iteration to carry out image filtering processing;

Performing image enhancement processing on an L component of a color space of a substrate image LAB to obtain an L component enhanced image;

And merging the a component filtered image, the b component filtered image and the L component enhanced image and converting RGB color space to determine a preprocessed substrate image.

5. The method for inspecting a transparent flexible substrate for an integrated circuit according to claim 4, wherein performing image enhancement processing on the L component of the color space of the substrate image LAB to obtain an L component enhanced image comprises:

acquiring pixel values of all pixel points of an L component of a color space of a substrate image LAB, marking the pixel values as background pixels when the pixel values are not lower than a preset pixel threshold value, and extracting a background low-illumination area;

Performing illumination image decomposition and reflection image decomposition on a background low-illumination area, performing gray enhancement processing on the illumination image, and performing filtering processing on the reflection image;

And combining the illumination image subjected to the gray enhancement processing with the reflection image subjected to the filtering processing to obtain an L component enhanced image.