An aoi particle indentation detection method and device and a storage medium

Abstract

An AOI particle indentation detection method, comprising: acquiring a first image of a PFC flat cable to be detected; generating a second image based on an AOI imaging algorithm and the first image, the second image comprising at least particle quantity, bubbles and area cracks; comparing the second image with preset image data based on an image comparison algorithm to generate a corresponding recognition result; acquiring the first image, generating the second image containing particle quantity, bubbles and area cracks based on the AOI imaging algorithm, comparing the second image with the preset image data based on the image comparison algorithm to generate the corresponding recognition result, detecting particle quantity, bubbles and area cracks, making the automatic detection process more comprehensive, making the detected product result more reliable, and thus making the production line finished product qualification rate and product straight-through rate higher, and the production cost lower.

Description

[Technical Field]

[0001] This application relates to the field of communication technology, and in particular to an AOI particle indentation detection method, apparatus and storage medium. [Background Technology]

[0002] AOI (Auto Optical Inspection) is an automated optical identification system used in the electronic industry for the visual inspection of circuit board assembly lines, replacing the previous manual visual inspection. Its basic principle is to use imaging technology to compare the object under test with a standard image to determine whether the object under test meets the standard. Therefore, the quality of AOI basically depends on its image resolution, imaging capability and image analysis technology.

[0003] With technological advancements, AOI technology has gradually been applied to the detection of particle indentations on screen ICs and PFC cables on screens. However, existing automatic particle indentation detection devices do not detect the number of particles, bubbles, or regional cracks; they only detect chipped corners or edges of ICs. The detection of particle number, bubbles, and regional cracks on cables is not comprehensive enough, resulting in lower reliability of the detected product results. It is difficult to ensure a high pass rate for finished products on the production line, leading to a decrease in first-pass yield and an increase in defect rate, which can easily increase production costs and reduce profits. [Summary of the Invention]

[0004] To address the issue that existing automated particle indentation detection devices are insufficient in detecting the number of particles, bubbles, and regional cracks on the production line, leading to lower reliability of the detected product results and difficulty in ensuring a high pass rate for finished products on the production line, resulting in a decrease in the first-pass yield, this invention proposes an AOI particle indentation detection method. This method involves acquiring a first image, generating a second image based on an AOI imaging algorithm, and finally generating the corresponding recognition result through an image comparison algorithm and preset image data.

[0005] The present invention proposes the following solution:

[0006] An AOI particle indentation detection method includes:

[0007] Acquire the first image of the PFC cable under test;

[0008] Based on the AOI imaging algorithm and the first image, a second image is generated, which includes at least the number of particles, bubbles, and regional cracks.

[0009] The second image and the preset image data are compared based on the image comparison algorithm to generate the corresponding recognition results.

[0010] The AOI particle indentation detection method described above, wherein the step of acquiring the first image of the PFC cable under test includes:

[0011] Based on the preset light source angle and light source intensity, the light reflected from the surface of the PFC cable under test is obtained;

[0012] A two-dimensional planar image is generated based on the light reflected from the surface of the PFC cable under test.

[0013] The first image is determined based on the two-dimensional planar image.

[0014] The AOI particle indentation detection method described above, wherein the step of acquiring the light reflected from the surface of the PFC cable under test according to a preset light source angle and light source intensity includes:

[0015] Obtain the installation angle and cable type of the PFC cable to be tested;

[0016] Determine the light source angle based on the installation angle of the PFC cable under test;

[0017] Determine the light transmittance and thickness of the PFC cable to be tested based on the cable type;

[0018] The light source intensity is determined based on the light transmittance and thickness.

[0019] The light source is used to illuminate the PFC cable under test according to the light source angle and light source intensity, and the light reflected from the surface of the PFC cable under test is obtained.

[0020] The AOI particle indentation detection method described above, wherein the step of generating a second image based on the AOI imaging algorithm and the first image, wherein the second image includes at least the particle number, bubbles, and regional cracks, includes:

[0021] Based on the AOI imaging algorithm, a third image is obtained by preprocessing the first image. The preprocessing includes at least denoising, smoothing, and enhancing the first image.

[0022] Based on threshold segmentation and edge detection algorithms, the number, size and shape of particles in the third image are determined.

[0023] Based on the Yolov3 algorithm and the first image, regional crack data and bubble data are generated.

[0024] The types of particles are determined based on the k-means++ algorithm, particle size, particle shape, regional crack data, and bubble data.

[0025] A second image is generated based on the first image and the classified particles.

[0026] The AOI particle indentation detection method described above, wherein the step of comparing the second image and preset image data based on the image comparison algorithm to generate the corresponding recognition result includes:

[0027] The particle anomaly threshold is determined by comparing the second image with the preset image data based on the image comparison algorithm.

[0028] Based on the particle anomaly threshold, a corresponding identification result is generated, which includes at least normal, critical anomaly, and anomaly.

[0029] An AOI particle indentation detection device, comprising:

[0030] The acquisition module is used to acquire the first image of the PFC cable under test.

[0031] The first generation module is used to generate a second image based on the AOI imaging algorithm and the first image. The second image includes at least the number of particles, bubbles, and regional cracks.

[0032] The second generation module is used to compare the second image with preset image data based on an image comparison algorithm to generate the corresponding recognition result.

[0033] The AOI particle indentation detection device described above, wherein the acquisition module includes:

[0034] The acquisition unit is used to acquire the light reflected from the surface of the PFC cable under test according to the preset light source angle and light source intensity.

[0035] The first generation unit is used to generate a two-dimensional planar image based on the light reflected from the surface of the PFC cable under test.

[0036] The first determining unit is used to determine the first image based on the two-dimensional planar image;

[0037] The acquisition unit includes:

[0038] The first acquisition subunit is used to acquire the installation angle and cable type of the PFC cable under test.

[0039] The first determining subunit is used to determine the light source angle based on the installation angle of the PFC cable under test;

[0040] The second determining subunit is used to determine the light transmittance and thickness of the PFC cable to be tested based on the cable type.

[0041] The third determining subunit is used to determine the light source intensity based on the light transmittance and thickness;

[0042] The second acquisition subunit is used to illuminate the PFC cable under test according to the light source angle and light source intensity, and acquire the light reflected from the surface of the PFC cable under test.

[0043] The first generation module includes:

[0044] The processing unit is used to preprocess the first image to obtain the third image based on the AOI imaging algorithm. The preprocessing includes at least denoising, smoothing and enhancing the first image.

[0045] The second determining unit is used to determine the number, size and shape of particles in the third image based on the threshold segmentation algorithm and the edge detection algorithm.

[0046] The second generation unit is used to generate regional crack data and bubble data based on the Yolov3 algorithm and the first image.

[0047] The third determining unit is used to determine the type of particle based on the k-means++ algorithm, particle size, particle shape, regional crack data, and bubble data.

[0048] The third generation unit is used to generate a second image based on the first image and the classified particles;

[0049] The second generation module includes:

[0050] The fourth determining unit is used to compare the second image with preset image data based on an image comparison algorithm to determine the particle anomaly threshold;

[0051] The fourth generation unit is used to generate corresponding recognition results based on the particle anomaly threshold, wherein the recognition results include at least normal, critical anomaly, and anomaly.

[0052] A computer-readable storage medium storing a computer program that, when executed by an AOI particle indentation detection device, implements the AOI particle indentation detection method as described above.

[0053] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the AOI particle indentation detection method as described above.

[0054] This invention acquires a first image, then generates a second image containing data on particle count, bubbles, and regional cracks based on an AOI imaging algorithm. Finally, it compares the second image with preset image data using an image comparison algorithm to generate corresponding recognition results. This process detects particle count, bubbles, and regional cracks, making the automatic detection process more comprehensive and the product results more reliable. This results in higher finished product qualification rate and first-pass yield on the production line, leading to lower production costs. [Attached Image Description]

[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a flowchart of the AOI particle indentation detection method according to the first embodiment of the present invention;

[0057] Figure 2 yes Figure 1 Detailed flowchart of step S11;

[0058] Figure 3 yes Figure 2 Detailed flowchart of step S111;

[0059] Figure 4 yes Figure 1 Detailed flowchart of step S12;

[0060] Figure 5 yes Figure 1 Detailed flowchart of step S13;

[0061] Figure 6 This is a flowchart of the AOI particle indentation detection method according to the second embodiment of the present invention;

[0062] Figure 7 This is a structural block diagram of the AOI particle indentation detection device according to the third embodiment of the present invention;

[0063] Figure 8 yes Figure 6 The detailed structural diagram of the module is obtained;

[0064] Figure 9 yes Figure 8 Obtain the detailed structural block diagram of the unit;

[0065] Figure 10 yes Figure 6 Detailed structural block diagram of the first generation module;

[0066] Figure 11 yes Figure 6 Detailed structural block diagram of the second generation module;

[0067] Figure 12 This is a structural block diagram of a computer device according to another embodiment of the present invention.

Detailed Implementation Methods

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Well-known modules, units, and their connections, links, communications, or operations are not shown or described in detail. Furthermore, the described features, architectures, or functions can be combined in any way in one or more embodiments. Those skilled in the art should understand that the various embodiments described below are only for illustrative purposes and not for limiting the scope of protection of the present invention. It is also readily understood that the modules, units, or processing methods in the various embodiments described herein and shown in the accompanying drawings can be combined and designed in various different configurations. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] The definitions of various terms or methods used in the following embodiments are, except where logically impossible, generally defined as broad concepts that can be implemented under the premise of the content disclosed in the embodiments. Under this understanding, all specific subordinate limitations of the terms or methods should be considered as part of the invention, and should not be narrowly interpreted or biased simply because the specification does not disclose such a specific limitation. For example, when the present invention refers to a cloud platform, it includes not only virtual network servers but also real physical devices, which not only have data storage capabilities but also data processing, intelligent analysis, and reasoning capabilities. Similarly, provided logically feasible, the order of steps in the method is flexible and varied, and all specific subordinate limitations within the broad concepts of various terms or methods fall within the scope of protection of this invention.

[0070] First embodiment:

[0071] Please refer to Figures 1 to 5 As shown, this embodiment proposes an AOI particle indentation detection method, including S11-S13, wherein:

[0072] S11. Obtain the first image of the PFC cable under test.

[0073] In this embodiment, before the detection begins, the CCD camera on the detection device acquires an image of the PFC cable under test. The image of the PFC cable under test includes the particle indentation image of the screen IC and the particle indentation image of the PFC cable on the screen. The two image information are then used to determine whether there are any abnormalities in the particles on the cable and the screen IC, making the detection process more accurate.

[0074] As a preferred option rather than a specific limitation, step S11 includes S111-S113, wherein:

[0075] S111. Obtain the light reflected from the surface of the PFC cable under test according to the preset light source angle and light source intensity.

[0076] Before the test begins, different light sources need to be determined according to different models of ribbon cables. This involves the angle and intensity of the light source to accurately obtain the light reflected from the surface of the PFC ribbon cable under test, thereby better obtaining the image information of the PFC ribbon cable under test and making the testing process more accurate.

[0077] As a preferred embodiment rather than a specific limitation, step S111 includes S1111-S1115, wherein:

[0078] S1111: Obtain the installation angle and cable model of the PFC cable to be tested.

[0079] This embodiment obtains the installation angle and model of the PFC cable under test, and then determines the light source. This allows for more accurate acquisition of the light reflected from the surface of the PFC cable under test, thus providing a better image of the PFC cable. Since the testing in this embodiment mainly targets screen cables and screen ICs, the installation angle is set for different screen sizes. Furthermore, during the placement process before testing, the screen angle may tilt due to inaccurate installation position of the device. In addition, different models of cables may have certain deviations in light transmittance or thickness. By determining these two factors, a better light source can be adjusted, making the testing process more stable.

[0080] S1112. Determine the light source angle based on the installation angle of the PFC cable to be tested.

[0081] This embodiment determines the position more accurately based on the actual installation angle of the PFC cable under test, including the rotation angle change in the plane, which can better determine the angle of the light source and make the mapped image clearer.

[0082] S1113. Determine the light transmittance and thickness of the PFC cable to be tested based on the cable model.

[0083] In this embodiment, based on the obtained cable model, the system pre-stores data on the PFC cables to be tested for each model, and selects the data on light transmittance and thickness. The light transmittance is compared with the emitted light source value and the light source value after transmission, and the formula is as follows:

[0084] S = (a / b) * 100%

[0085] Where S represents light transmittance, a represents the light source value after transmission, and b represents the emitted light source value.

[0086] S1114. Determine the light source intensity based on the light transmittance and thickness.

[0087] This embodiment determines the required light source intensity by considering light transmittance and thickness, so that the intensity of the reflected light is appropriate, thereby making the acquired image clear enough and improving the automatic detection effect.

[0088] S1115. Illuminate the PFC cable under test according to the light source angle and light source intensity, and obtain the light reflected from the surface of the PFC cable under test.

[0089] After determining the light source angle and intensity in this embodiment, the PFC cable under test is illuminated according to the preset light source intensity and corresponding angle, so that the PFC cable under test reflects light into the CCD camera. The CCD camera collects the corresponding image data, thereby better obtaining the image information of the PFC cable under test and making the detection process more accurate.

[0090] S112. Generate a two-dimensional planar image based on the light reflected from the surface of the PFC cable under test.

[0091] This implementation obtains the light reflected from the surface of the PFC cable under test through the above process, analyzes and converts it according to the light conditions, and constructs a two-dimensional planar image based on the different shadow surfaces, which can make the first image data clearer.

[0092] S113. Determine the first image based on the two-dimensional planar image.

[0093] In this embodiment, a first image is generated based on a predetermined two-dimensional planar image through computer data conversion, so as to make the detected image more accurate and reliable.

[0094] S12. Based on the AOI imaging algorithm and the first image, generate a second image, the second image including at least the number of particles, bubbles and regional cracks.

[0095] This embodiment uses an AOI imaging algorithm to disassemble and analyze the first image to generate a second image that is easy to process. The image data includes at least the number of particles, bubbles, and regional cracks, thereby determining whether there are any abnormalities in the PFC cable under test, so that the subsequent comparison effect is more obvious and the test results are more comprehensive and accurate.

[0096] The particle AOI involved in this embodiment is a particle detection technology that uses optical principles. Its algorithm is mainly used to process images for denoising, smoothing, enhancement, etc., and then to detect particles. Through optical imaging and image processing technology, the image is processed and analyzed to achieve particle detection and classification.

[0097] As a preferred option rather than a specific limitation, step S12 includes S121-S125, wherein:

[0098] S121. Based on the AOI imaging algorithm, the first image is preprocessed to obtain the third image. The preprocessing includes at least denoising, smoothing and enhancing the first image.

[0099] In this embodiment, the first step is to perform noise reduction, smoothing and enhancement processing on the image using the AOI imaging algorithm to make the image features more obvious. The data calculated by the acquired third image is more accurate and the detection effect is better.

[0100] S122. Based on threshold segmentation algorithm and edge detection algorithm, determine the number, size and shape of particles in the third image.

[0101] This embodiment uses a threshold segmentation algorithm to segment the third image into regions, determine the number of particles in different regions, and then uses an edge detection algorithm to detect the edges of the particles, finally obtaining the particle size and shape in the third image to make the detection effect more accurate.

[0102] S123. Based on the Yolov3 algorithm and the first image, generate regional crack data and bubble data.

[0103] This embodiment uses the Yolov3 algorithm to detect regional crack data and bubble data targets in the first image. It can stably and efficiently determine regional crack data and bubble data. The bubble refers to some wear caused by the compression process of the particles, which is like a bubble. The regional crack refers to the cracks and other data on the PFC cable to be tested.

[0104] S124. Based on the k-means++ algorithm, particle size, particle shape, regional crack data, and bubble data, determine the type of particle.

[0105] This embodiment uses the k-means++ algorithm to cluster data such as particle size, particle shape, regional crack data, and bubble data, distinguishing different particle sizes, particle shapes, such as circular and elliptical shapes, and identifying whether bubbles exist in the image, which are generated during particle compression.

[0106] S125. Generate a second image based on the first image and the classified particles.

[0107] In this embodiment, the second image is generated by fusing the first image data and the classification results. Because the second image data has been processed, it can be analyzed and identified better, and the generated result is more accurate and reliable.

[0108] S13. Based on the image comparison algorithm, compare the second image with the preset image data to generate the corresponding recognition result.

[0109] This embodiment uses an image comparison algorithm to compare the second image with preset image data, thereby determining whether there is an abnormality in the particles on the PFC line under test. The determination process is relatively accurate and reliable.

[0110] The image matching algorithm described in this embodiment uses pixel-level matching, centroid matching, projection matching, or block-level matching algorithms for comparison and determination. This type of algorithm treats the image as a matrix, where each element is a color value composed of three RGB parameters, ranging from 0 to 255. Of course, image representation is not limited to RGB; other types will not be detailed here. Since the range of 0 to 255 is too large, the image needs to be reduced in dimensionality—binarized. Binarization transforms the image into only black and white, which can be achieved using the OTSU algorithm. We then represent black with 1 and white with 0, resulting in a matrix consisting only of the numbers 0 and 1.

[0111] As a preferred option rather than a specific limitation, step S13 includes S131-S132, wherein:

[0112] S131. Based on the image comparison algorithm, compare the second image with the preset image data to determine the particle anomaly threshold.

[0113] In order to determine particle anomalies, this embodiment sets a threshold range. Anomalies within the range are considered normal, while those exceeding the threshold are considered abnormal, in order to avoid misjudgment and make the testing process more accurate.

[0114] S132. Generate corresponding identification results based on particle anomaly thresholds, wherein the identification results include at least normal, critical anomaly, and anomaly.

[0115] This embodiment sets a certain particle threshold and then divides the particle anomaly threshold data into three parts: normal, critically abnormal, and abnormal. Data within the threshold range is considered normal and is identified as normal results. Data just above the threshold line is considered critical and is identified as critical results. Data exceeding the threshold range is considered abnormal and is identified as abnormal results. By differentiating the results, the effect is better.

[0116] This embodiment acquires a first image, then generates a second image containing data on particle count, bubbles, and regional cracks based on an AOI imaging algorithm. Finally, it compares the second image with preset image data using an image comparison algorithm to generate corresponding recognition results. This process detects particle count, bubbles, and regional cracks, making the automatic detection process more comprehensive and the product results more reliable. This results in higher finished product qualification rate and first-pass yield on the production line, leading to lower production costs.

[0117] Second embodiment:

[0118] Please refer to Figure 6 As shown, this embodiment also provides an AOI particle indentation detection method, which can be executed after S11 and before S12 of the first embodiment described above. It is an alternative implementation of the first embodiment, and steps S12 and thereafter are also synchronously replaced. The AOI particle indentation detection method includes S31-S33, wherein:

[0119] S201. Obtain the installation angle and cable model of the PFC cable to be tested.

[0120] This embodiment obtains the installation angle and model of the PFC cable under test, and then determines the light source. This allows for more accurate acquisition of the light reflected from the surface of the PFC cable under test, thus providing a better image of the PFC cable. Since the testing in this embodiment mainly targets screen cables and screen ICs, the installation angle is set for different screen sizes. Furthermore, during the placement process before testing, the screen angle may tilt due to inaccurate installation position of the device. In addition, different models of cables may have certain deviations in light transmittance or thickness. By determining these two factors, a better light source can be adjusted, making the testing process more stable.

[0121] S202. Determine the light source angle based on the installation angle of the PFC cable to be tested.

[0122] This embodiment determines the position more accurately based on the actual installation angle of the PFC cable under test, including the rotation angle change in the plane, which can better determine the angle of the light source and make the mapped image clearer.

[0123] S203. Determine the light transmittance and thickness of the PFC cable to be tested based on the cable model.

[0124] In this embodiment, based on the obtained cable model, the system pre-stores data on the PFC cables to be tested for each model, and selects the data on light transmittance and thickness. The light transmittance is compared with the emitted light source value and the light source value after transmission, and the formula is as follows:

[0125] S = (a / b) * 100%

[0126] Where S represents light transmittance, a represents the light source value after transmission, and b represents the emitted light source value.

[0127] S204. Determine the light source intensity based on the light transmittance and thickness.

[0128] This embodiment determines the required light source intensity by considering light transmittance and thickness, so that the intensity of the reflected light is appropriate, thereby making the acquired image clear enough and improving the automatic detection effect.

[0129] S205. Illuminate the PFC cable under test according to the light source angle and light source intensity, and obtain the light reflected from the surface of the PFC cable under test.

[0130] After determining the light source angle and intensity in this embodiment, the PFC cable under test is illuminated according to the preset light source intensity and corresponding angle, so that the PFC cable under test reflects light into the CCD camera. The CCD camera collects the corresponding image data, thereby better obtaining the image information of the PFC cable under test and making the detection process more accurate.

[0131] S206. Based on the AOI imaging algorithm, the first image is preprocessed to obtain the third image. The preprocessing includes at least denoising, smoothing and enhancing the first image.

[0132] This embodiment uses an AOI imaging algorithm to disassemble and analyze the first image to generate a second image that is easy to process. The image data includes at least the number of particles, bubbles, and regional cracks, thereby determining whether there are any abnormalities in the PFC cable under test, so that the subsequent comparison effect is more obvious and the test results are more comprehensive and accurate.

[0133] The particle AOI involved in this embodiment is a particle detection technology that uses optical principles. Its algorithm is mainly used to process images for denoising, smoothing, enhancement, etc., and then to detect particles. Through optical imaging and image processing technology, the image is processed and analyzed to achieve particle detection and classification.

[0134] In this embodiment, the first step is to perform noise reduction, smoothing and enhancement processing on the image using the AOI imaging algorithm to make the image features more obvious. The data calculated by the acquired third image is more accurate and the detection effect is better.

[0135] S207. Based on threshold segmentation algorithm and edge detection algorithm, determine the number, size and shape of particles in the third image.

[0136] This embodiment uses a threshold segmentation algorithm to segment the third image into regions, determine the number of particles in different regions, and then uses an edge detection algorithm to detect the edges of the particles, finally obtaining the particle size and shape in the third image to make the detection effect more accurate.

[0137] S208. Based on the Yolov3 algorithm and the first image, regional crack data and bubble data are generated.

[0138] This embodiment uses the Yolov3 algorithm to detect regional crack data and bubble data targets in the first image. It can stably and efficiently determine regional crack data and bubble data. The bubble refers to some wear caused by the compression process of the particles, which is like a bubble. The regional crack refers to the cracks and other data on the PFC cable to be tested.

[0139] S209. Based on the k-means++ algorithm, particle size, particle shape, regional crack data, and bubble data, determine the type of particle.

[0140] This embodiment uses the k-means++ algorithm to cluster data such as particle size, particle shape, regional crack data, and bubble data, distinguishing different particle sizes, particle shapes, such as circular and elliptical shapes, and identifying whether bubbles exist in the image, which are generated during particle compression.

[0141] S210. Generate a second image based on the first image and the classified particles.

[0142] In this embodiment, the second image is generated by fusing the first image data and the classification results. Because the second image data has been processed, it can be analyzed and identified better, and the generated result is more accurate and reliable.

[0143] S211. Based on the image comparison algorithm, compare the second image with the preset image data to determine the particle anomaly threshold.

[0144] This embodiment uses an image comparison algorithm to compare the second image with preset image data to determine whether there is an abnormality in the particles on the PFC cable under test. A threshold range is set. Abnormalities within the range are considered normal, while those exceeding the threshold are considered abnormal, in order to avoid misjudgment. The testing process is more accurate, and the determination process is more precise and reliable.

[0145] The image matching algorithm described in this embodiment uses pixel-level matching, centroid matching, projection matching, or block-level matching algorithms for comparison and determination. This type of algorithm treats the image as a matrix, where each element is a color value composed of three RGB parameters, ranging from 0 to 255. Of course, image representation is not limited to RGB; other types will not be detailed here. Since the range of 0 to 255 is too large, the image needs to be reduced in dimensionality—binarized. Binarization transforms the image into only black and white, which can be achieved using the OTSU algorithm. We then represent black with 1 and white with 0, resulting in a matrix consisting only of the numbers 0 and 1.

[0146] S212. Generate corresponding identification results based on particle anomaly thresholds, wherein the identification results include at least normal, critical anomaly, and anomaly.

[0147] This embodiment sets a certain particle threshold and then divides the particle anomaly threshold data into three parts: normal, critically abnormal, and abnormal. Data within the threshold range is considered normal and is identified as normal results. Data just above the threshold line is considered critical and is identified as critical results. Data exceeding the threshold range is considered abnormal and is identified as abnormal results. By differentiating the results, the effect is better.

[0148] This embodiment acquires a first image, then generates a second image containing data on particle count, bubbles, and regional cracks based on an AOI imaging algorithm. Finally, it compares the second image with preset image data using an image comparison algorithm to generate corresponding recognition results. This process detects particle count, bubbles, and regional cracks, making the automatic detection process more comprehensive and the product results more reliable. This results in higher finished product qualification rate and first-pass yield on the production line, leading to lower production costs.

[0149] Third embodiment:

[0150] Please refer to Figures 7 to 11 As shown, this embodiment proposes an AOI particle indentation detection device 100, including an acquisition module 110, a first generation module 120, and a second generation module 130, wherein:

[0151] The acquisition module 110 is connected to the first generation module 120 and is used to acquire the first image of the PFC cable under test.

[0152] As a preferred embodiment rather than a specific limitation, the acquisition module 110 includes an acquisition unit 111, a first generation unit 112, and a first determination unit 113, wherein:

[0153] The acquisition unit 111 is connected to the first generation unit 112 and is used to acquire the light reflected from the surface of the PFC cable under test according to the preset light source angle and light source intensity.

[0154] The acquisition unit 111 includes a first acquisition subunit 1111, a first determination subunit 1112, a second determination subunit 1113, a third determination subunit 1114, and a second acquisition subunit 1115, wherein:

[0155] The first acquisition subunit 1111 and the first determination subunit 1112 are connected to acquire the installation angle and cable model of the PFC cable to be tested.

[0156] The first determining subunit 1112 and the second determining subunit 1113 are connected to determine the light source angle based on the installation angle of the PFC cable to be tested.

[0157] The second determining subunit 1113 and the third determining subunit 1114 are connected and used to determine the light transmittance and thickness of the PFC cable to be tested according to the cable model.

[0158] The third determining subunit 1114 is connected to the second obtaining subunit 1115 and is used to determine the light source intensity based on the light transmittance and thickness.

[0159] The second acquisition subunit 1115 is used to illuminate the PFC cable under test according to the light source angle and light source intensity, and acquire the light reflected from the surface of the PFC cable under test.

[0160] The first generating unit 112 is connected to the first determining unit 113 and is used to generate a two-dimensional planar image based on the light reflected from the surface of the PFC cable to be tested.

[0161] The first determining unit 113 is used to determine the first image based on the two-dimensional planar image.

[0162] The first generation module 120 is connected to the second generation module 130 and is used to generate a second image based on the AOI imaging algorithm and the first image. The second image includes at least the number of particles, bubbles, and regional cracks.

[0163] As a preferred embodiment rather than a specific limitation, the first generation module 120 includes a processing unit 121, a second determining unit 122, a second generation unit 123, a third determining unit 124, and a third generation unit 125, wherein:

[0164] The processing unit 121 is connected to the second determining unit 122 and is used to preprocess the first image to obtain the third image based on the AOI imaging algorithm. The preprocessing includes at least denoising, smoothing and enhancing the first image.

[0165] The second determining unit 122 is connected to the second generating unit 123 and is used to determine the number, size and shape of particles in the third image based on the threshold segmentation algorithm and the edge detection algorithm.

[0166] The second generation unit 123 is connected to the third determination unit 124 and is used to generate regional crack data and bubble data based on the YOLOv3 algorithm and the first image.

[0167] The third determining unit 124 is connected to the third generating unit 125 and is used to determine the type of particle based on the k-means++ algorithm, particle size, particle shape, regional crack data and bubble data.

[0168] The third generation unit 125 is used to generate a second image based on the first image and the classified particles.

[0169] The second generation module 130 is used to compare the second image and the preset image data based on the image comparison algorithm to generate the corresponding recognition result.

[0170] As a preferred embodiment rather than a specific limitation, the second generation module 130 includes a fourth determining unit 131 and a fourth generation unit 132, wherein:

[0171] The fourth determining unit 131 is connected to the fourth generating unit 132 and is used to compare the second image with preset image data based on the image comparison algorithm to determine the particle anomaly threshold.

[0172] The fourth generation unit 132 is used to generate corresponding recognition results based on the particle anomaly threshold, wherein the recognition results include at least normal, critical anomaly, and anomaly.

[0173] This embodiment acquires a first image, then generates a second image containing data on particle count, bubbles, and regional cracks based on an AOI imaging algorithm. Finally, it compares the second image with preset image data using an image comparison algorithm to generate corresponding recognition results. This process detects particle count, bubbles, and regional cracks, making the automatic detection process more comprehensive and the product results more reliable. This results in higher finished product qualification rate and first-pass yield on the production line, leading to lower production costs.

[0174] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0175] This invention also provides a computer storage medium storing a computer program that, when executed by a processor, implements an AOI particle indentation detection method as described in the above embodiments.

[0176] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of each of the above embodiments of the AOI particle indentation detection method. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous RAM (SDRAM), dual data rate SDR (DDR SDRAM), enhanced SDR (ESDRAM), synchronous link RAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0177] Alternatively, if the integrated units of the present invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the parts that contribute to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, terminal, or network device, etc.) to execute all or part of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, RAM, ROM, magnetic disks, or optical disks. Corresponding to the aforementioned computer storage medium, one embodiment also provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements an AOI particle indentation detection method as described in the above embodiments.

[0178] This computer device can be a terminal, and its internal structure diagram can be as follows: Figure 12 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements an AOI particle indentation detection method. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0179] This embodiment acquires a first image, then generates a second image containing data on particle count, bubbles, and regional cracks based on an AOI imaging algorithm. Finally, it compares the second image with preset image data using an image comparison algorithm to generate corresponding recognition results. This process detects particle count, bubbles, and regional cracks, making the automatic detection process more comprehensive and the product results more reliable. This results in higher finished product qualification rate and first-pass yield on the production line, leading to lower production costs.

[0180] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0181] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. An AOI particle indentation detection method, characterized in that, include: Acquire the first image of the PFC cable under test; Based on the AOI imaging algorithm and the first image, a second image is generated, which includes at least the number of particles, bubbles, and regional cracks. Based on the image comparison algorithm, the second image and the preset image data are compared to generate the corresponding recognition results; The step of acquiring the first image of the PFC cable under test includes: Based on the preset light source angle and light source intensity, the light reflected from the surface of the PFC cable under test is obtained; A two-dimensional planar image is generated based on the light reflected from the surface of the PFC cable under test. The first image is determined based on the two-dimensional planar image; The step of generating a second image based on the AOI imaging algorithm and the first image, wherein the second image includes at least the particle number, bubbles, and regional cracks, includes: Based on the AOI imaging algorithm, a third image is obtained by preprocessing the first image. The preprocessing includes at least denoising, smoothing, and enhancing the first image. Based on threshold segmentation and edge detection algorithms, the number, size and shape of particles in the third image are determined. Based on the Yolov3 algorithm and the first image, regional crack data and bubble data are generated. The types of particles are determined based on the k-means++ algorithm, particle size, particle shape, regional crack data, and bubble data. A second image is generated based on the first image and the classified particles.

2. The AOI particle indentation detection method according to claim 1, characterized in that, The step of obtaining the light reflected from the surface of the PFC cable under test according to the preset light source angle and light source intensity includes: Obtain the installation angle and cable type of the PFC cable to be tested; Determine the light source angle based on the installation angle of the PFC cable under test; Determine the light transmittance and thickness of the PFC cable to be tested based on the cable type; The light source intensity is determined based on the light transmittance and thickness. The light source is used to illuminate the PFC cable under test according to the light source angle and light source intensity, and the light reflected from the surface of the PFC cable under test is obtained.

3. The AOI particle indentation detection method according to claim 1, characterized in that, The step of comparing the second image and preset image data based on the image comparison algorithm to generate the corresponding recognition result includes: The particle anomaly threshold is determined by comparing the second image with the preset image data based on the image comparison algorithm. Based on the particle anomaly threshold, a corresponding identification result is generated, which includes at least normal, critical anomaly, and anomaly.

4. An AOI particle indentation detection device, characterized in that, include: The acquisition module is used to acquire the first image of the PFC cable under test. The first generation module is used to generate a second image based on the AOI imaging algorithm and the first image. The second image includes at least the number of particles, bubbles, and regional cracks. The second generation module is used to compare the second image with the preset image data based on the image comparison algorithm and generate the corresponding recognition result. The acquisition module is further configured to acquire the light reflected from the surface of the PFC cable under test according to a preset light source angle and light source intensity; generate a two-dimensional planar image based on the light reflected from the surface of the PFC cable under test; and determine a first image based on the two-dimensional planar image. The first generation module is further configured to preprocess the first image to obtain a third image based on an AOI imaging algorithm. The preprocessing includes at least denoising, smoothing, and enhancing the first image; determining the number, size, and shape of particles in the third image based on a threshold segmentation algorithm and an edge detection algorithm; generating regional crack data and bubble data based on the Yolov3 algorithm and the first image; determining the type of particles based on the k-means++ algorithm, particle size, particle shape, regional crack data, and bubble data; and generating a second image based on the first image and the classified particles. The acquisition module includes: The acquisition unit is used to acquire the light reflected from the surface of the PFC cable under test according to the preset light source angle and light source intensity. The first generation unit is used to generate a two-dimensional planar image based on the light reflected from the surface of the PFC cable under test. The first determining unit is used to determine the first image based on the two-dimensional planar image; The acquisition unit includes: The first acquisition subunit is used to acquire the installation angle and cable type of the PFC cable under test. The first determining subunit is used to determine the light source angle based on the installation angle of the PFC cable under test; The second determining subunit is used to determine the light transmittance and thickness of the PFC cable to be tested based on the cable type. The third determining subunit is used to determine the light source intensity based on the light transmittance and thickness; The second acquisition subunit is used to illuminate the PFC cable under test according to the light source angle and light source intensity, and acquire the light reflected from the surface of the PFC cable under test. The first generation module includes: The processing unit is used to preprocess the first image to obtain the third image based on the AOI imaging algorithm. The preprocessing includes at least denoising, smoothing and enhancing the first image. The second determining unit is used to determine the number, size and shape of particles in the third image based on the threshold segmentation algorithm and the edge detection algorithm. The second generation unit is used to generate regional crack data and bubble data based on the Yolov3 algorithm and the first image. The third determining unit is used to determine the type of particle based on the k-means++ algorithm, particle size, particle shape, regional crack data, and bubble data. The third generation unit is used to generate a second image based on the first image and the classified particles; The second generation module includes: The fourth determining unit is used to compare the second image with preset image data based on an image comparison algorithm to determine the particle anomaly threshold; The fourth generation unit is used to generate corresponding recognition results based on the particle anomaly threshold, wherein the recognition results include at least normal, critical anomaly, and anomaly.

5. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by the AOI particle indentation detection device, implements the AOI particle indentation detection method as described in any one of claims 1-3.

6. A computer device, characterized in that, The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the AOI particle indentation detection method as described in any one of claims 1-3.

Images