Non-defective product prediction method and system for computer industrial control mainboard

Abstract

The invention provides a good product prediction method and system for a computer industrial control mainboard, and the method comprises the steps: obtaining all welding spot images through collecting the images of all welding spots when a to-be-detected computer industrial control mainboard operates in detection equipment, and determining the pixel contribution degree of the welding spots in each welding spot image; determining a global feature vector and a local feature vector of the welding spot based on the contribution degree of each pixel, and fusing the global feature vector and the local feature vector into a defect description quantity of the welding spot; determining the defect similarity coefficient of each welding spot image and the historical mainboard welding spot defect sample set, and determining the welding spot damage degree of the to-be-detected computer industrial control mainboard according to all the defect similarity coefficients and the defect description quantity; and predicting the use quality of the to-be-detected computer industrial control mainboard based on the welding spot damage degree, and marking good products. By adopting the scheme of the invention, the hidden trouble fault of the welding spot defect of the computer industrial control mainboard can be detected in advance, so that the credible range of good product detection of the computer industrial control mainboard is expanded.

Description

Technical Field

[0001] The present application relates to the technical field of computer industrial control motherboard detection, and more specifically, to a method and system for predicting good products for computer industrial control motherboards. Background Art

[0002] Computer industrial control motherboards, namely industrial control computer motherboards, are computer motherboards specially designed for use in industrial environments. Computer industrial control motherboards have higher reliability, stability and durability, and can work stably in harsh environments. Therefore, computer industrial control motherboards are widely used in industrial automation, transportation, energy, medical equipment and other fields. Due to its application in critical tasks and harsh environments, the quality requirements for computer industrial control motherboards are very high, so good product testing is particularly important.

[0003] The quality inspection of computer industrial control motherboards is a key link to ensure the quality of computer industrial control motherboards. The components on computer industrial control motherboards are connected by welding. There will always be welding problems in the welding process, such as cold solder joints, too much or too little solder joints, solder joint bridging and solder joint coplanarity. Therefore, the quality inspection of computer industrial control motherboards requires testing the internal welding quality, checking the integrity and consistency of the solder joints, and discovering potential manufacturing defects. However, in the prior art, the quality inspection of existing computer industrial control motherboards only determines that the computer industrial control motherboard is a good product if it can conduct electricity normally during the inspection. However, after long-term use or when the computer industrial control motherboard is violently shaken, a small amount of soldered parts may be damaged. The solder joints are prone to breakage, which may cause some pins to become skewed or even lead to faulty overlap with other pins, resulting in short circuit or open circuit in the computer industrial control motherboard during use, leading to hidden faults in the computer industrial control motherboard after installation and use. The existing good product detection equipment often relies on manual observation and judgment, which makes it difficult to detect internal defects and surface defects of solder joints. It is inefficient and prone to misjudgment, resulting in the outflow of poor computer industrial control motherboards with potential solder joint defects, which cause faults to occur after use, thus causing greater trouble to users. Therefore, how to achieve pre-detection of hidden faults of solder joint defects in computer industrial control motherboards, thereby increasing the credible range of good product detection of computer industrial control motherboards, has become a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides a method and system for predicting good products of computer industrial control mainboards, which can realize pre-detection of hidden faults of solder joint defects of computer industrial control mainboards, thereby increasing the credible range of good product detection of computer industrial control mainboards.

[0005] In a first aspect, the present application provides a method for predicting good products of a computer industrial control motherboard, comprising the following steps: Collect images corresponding to each solder joint of the industrial control motherboard of the computer to be tested when it is running in the testing equipment, obtain all solder joint images, and then determine the pixel contribution of the solder joint in each solder joint image; Based on the pixel contribution of each solder joint, the pixel points in all solder joint images are feature decomposed to obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, and the global feature vector and the local feature vector are merged into the defect description of the solder joint in the industrial control motherboard of the computer to be detected; Obtain a historical motherboard solder joint defect sample set, perform similarity matching on each solder joint image with the historical motherboard solder joint defect sample set, obtain a defect similarity coefficient between each solder joint image and the historical motherboard solder joint defect sample set, and determine the solder joint damage degree of the industrial control motherboard of the computer to be tested during operation based on all defect similarity coefficients and the defect description amount; The use quality of the industrial control mainboard of the computer to be inspected is predicted based on the degree of damage of the solder joints, and then the industrial control mainboard of the computer to be inspected is marked as good products according to the prediction result.

[0006] In some embodiments, determining the pixel contribution of each solder joint in each solder joint image specifically includes: Select a solder joint image; Based on a preset determination pixel value, each pixel in the solder joint image is divided into a defective pixel and a normal pixel; Determine the pixel contribution of the solder joint in the solder joint image based on all defective pixels and all normal pixels; Repeat the above steps to obtain the pixel contributions of the solder joints in the remaining solder joint images.

[0007] In some embodiments, the pixel points in all the solder joint images are feature decomposed based on the pixel contribution of each solder joint to obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, specifically including: The global feature space and local feature space of the solder joint defects in the computer industrial control motherboard to be inspected are determined by the pixel points in each solder joint image and the pixel contribution of each solder joint; Decomposing the global feature space to obtain the global feature vector of the solder joints of the industrial control motherboard of the computer to be detected on the global defect; The local feature space is decomposed to obtain the local feature vector of the solder joint on the local defect of the industrial control mainboard of the computer to be detected.

[0008] In some embodiments, performing similarity matching between each solder joint image and the historical motherboard solder joint defect sample set to obtain a defect similarity coefficient between each solder joint image and the historical motherboard solder joint defect sample set specifically includes: Select a solder joint image; Determine a matching value between the solder joint image and each defective solder joint sample image in the historical mainboard solder joint defect sample set; Determine the defect similarity coefficient between the solder joint image and the historical motherboard solder joint defect sample set based on all matching values; Repeat the above steps to obtain the defect similarity coefficients between the remaining solder joint images and the historical motherboard solder joint defect sample set.

[0009] In some embodiments, determining the degree of solder joint damage of the industrial control motherboard of the computer to be inspected during operation by using all defect similarity coefficients and the defect description quantity specifically includes: Determine the comprehensive matching degree between the industrial control motherboard of the computer to be tested and the solder joint defect sample set of the historical motherboard according to all defect similarity coefficients; The degree of damage to the solder joints of the industrial control mainboard of the computer to be inspected during operation is determined by the comprehensive matching degree and the defect description quantity.

[0010] In some embodiments, predicting the use quality of the industrial control motherboard of the computer to be inspected based on the degree of damage of the solder joints, and then marking the industrial control motherboard of the computer to be inspected as a good product according to the prediction result specifically includes: Preset the good product judgment threshold of the industrial control motherboard of the computer to be tested; When the solder joint damage degree is greater than the good product judgment threshold, it is predicted that the use quality of the industrial control motherboard of the computer to be tested is poor, and the industrial control motherboard of the computer to be tested is marked as a defective product; When the solder joint damage degree is less than or equal to the good product determination threshold, it is predicted that the use quality of the industrial control mainboard of the computer to be tested is good, and the industrial control mainboard of the computer to be tested is marked as a good product.

[0011] In some embodiments, an image acquisition device in a computer industrial control mainboard detection device is used to acquire images corresponding to various welding points when the computer industrial control mainboard to be detected is running in the detection device.

[0012] In a second aspect, the present application provides a good product prediction system for a computer industrial control motherboard, comprising: An acquisition module is used to acquire images corresponding to each solder joint when the industrial control motherboard of the computer to be tested is running in the testing device, obtain all solder joint images, and then determine the pixel contribution of the solder joint in each solder joint image; The processing module is further used to perform feature decomposition on the pixels in all solder joint images based on the pixel contribution of each solder joint, obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, and fuse the global feature vector and the local feature vector into a defect description of the solder joint in the industrial control motherboard of the computer to be detected; The processing module is also used to obtain a historical motherboard solder joint defect sample set, perform similarity matching between each solder joint image and the historical motherboard solder joint defect sample set, obtain a defect similarity coefficient between each solder joint image and the historical motherboard solder joint defect sample set, and determine the solder joint damage degree of the industrial control motherboard of the computer to be tested during operation based on all defect similarity coefficients and the defect description amount; The execution module is used to predict the use quality of the industrial control motherboard of the computer to be tested based on the solder joint damage degree, and then mark the industrial control motherboard of the computer to be tested as a good product according to the prediction result.

[0013] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-mentioned good product prediction method for a computer industrial control motherboard.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium, in which instructions or codes are stored. When the instructions or codes are run on a computer, the computer implements the above-mentioned good product prediction method for a computer industrial control motherboard when executing the computer.

[0015] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects: In the present application, by collecting images corresponding to each solder joint of the industrial control mainboard of the computer to be tested when it is running in the testing equipment, all the solder joint images are obtained, and then the pixel contribution of the solder joint in each solder joint image is determined; based on the pixel contribution of each solder joint, the pixel points in all the solder joint images are feature decomposed to obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, and the global feature vector and the local feature vector are merged into a defect description of the solder joint in the industrial control mainboard of the computer to be tested; a historical mainboard solder joint defect sample set is obtained, and each solder joint image is similarly matched with the historical mainboard solder joint defect sample set to obtain the defect similarity coefficient of each solder joint image and the historical mainboard solder joint defect sample set, and the solder joint damage degree of the industrial control mainboard of the computer to be tested during operation is determined by all the defect similarity coefficients and the defect description; based on the solder joint damage degree, the use quality of the industrial control mainboard of the computer to be tested is predicted, and then the industrial control mainboard of the computer to be tested is marked as good according to the prediction result.

[0016] It can be seen that in the present application, the degree of solder joint damage generated by the industrial control mainboard of the computer to be detected during operation is determined by all defect similarity coefficients and the defect description quantity; wherein, since the global feature vector reflects the defect situation of the solder joints of the industrial control mainboard of the computer to be detected on the global defects, and the local feature vector reflects the defect situation of the solder joints of the industrial control mainboard of the computer to be detected on the local defects, the global feature vector and the local feature vector can be fused to obtain the defect degree of the solder joints in the industrial control mainboard of the computer to be detected (i.e., the defect description quantity); furthermore, the degree of similarity between the solder joint image and the defect solder joint sample image in the historical mainboard solder joint defect sample set is determined to determine the degree of matching of each solder joint image with the historical mainboard solder joint defect sample set (i.e., the defect similarity). coefficient); and then, the degree of damage to the solder joints of the industrial control mainboard of the computer to be tested during operation is determined by all defect similarity coefficients and the defect description quantity, and the severity of the solder joint defects of the industrial control mainboard of the computer to be tested during operation is determined, thereby effectively making a pre-judgment of the hidden dangers of the solder joint defects of the industrial control mainboard of the computer to be tested; finally, the use quality of the industrial control mainboard of the computer to be tested is predicted based on the solder joint damage degree, and then the industrial control mainboard of the computer to be tested is marked as good products according to the prediction result, thereby completing the good product detection of the industrial control mainboard of the computer to be tested; in summary, the scheme realizes the pre-detection of the hidden dangers of the solder joint defects of the industrial control mainboard of the computer, thereby increasing the credible range of the good product detection of the industrial control mainboard of the computer. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0018] Figure 1 is an exemplary flow chart of a method for predicting good products of a computer industrial control motherboard according to some embodiments of the present application; Figure 2 is an exemplary flow chart of determining pixel contribution according to some embodiments of the present application; Figure 3 is an exemplary flow chart for determining a defect description amount according to some embodiments of the present application; Figure 4 It is a structural schematic diagram of a good product prediction system for a computer industrial control motherboard according to some embodiments of the present application; Figure 5 It is a structural schematic diagram of a computer device for implementing a method for predicting good products for a computer industrial control motherboard according to some embodiments of the present application. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0020] The embodiment of the present application provides a method and system for predicting good products of a computer industrial control motherboard. The core of the method and system is to start the detection equipment of the computer industrial control motherboard, and then connect the computer industrial control motherboard to be detected to the detection equipment; collect images corresponding to each solder joint when the computer industrial control motherboard to be detected is running in the detection equipment, obtain all solder joint images, and then determine the pixel contribution of the solder joint in each solder joint image; perform feature decomposition on the pixel points in all solder joint images based on the pixel contribution of each solder joint, obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, and fuse the global feature vector and the local feature vector into the image of the computer industrial control motherboard to be detected. The defect description of the solder joints in the board; obtaining a historical mainboard solder joint defect sample set, performing similarity matching on each solder joint image with the historical mainboard solder joint defect sample set, obtaining a defect similarity coefficient between each solder joint image and the historical mainboard solder joint defect sample set, and determining the solder joint damage degree of the computer industrial control mainboard to be tested during operation according to all defect similarity coefficients and the defect description; predicting the use quality of the computer industrial control mainboard to be tested based on the solder joint damage degree, and then marking the computer industrial control mainboard to be tested as a good product according to the prediction result; pre-detection of hidden dangers of computer industrial control mainboard solder joint defects can be achieved, thereby increasing the credible range of good product detection of computer industrial control mainboards.

[0021] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of a method for predicting good products for computer industrial control motherboards according to some embodiments of the present application.

[0022] In the specific implementation, it is necessary to first start the detection equipment of the computer industrial control motherboard, and then connect the industrial control motherboard of the computer to be detected to the detection equipment, that is: first, start the detection equipment of the computer industrial control motherboard and perform self-test and initialization, including hardware self-test, sensor calibration, image acquisition device adjustment and other processes to confirm that the various subsystems of the detection equipment (such as camera, sensor, image processing unit, etc.) are in normal state, and then, the interface (such as power interface, data interface, etc.) on the industrial control motherboard of the computer to be detected is matched and connected with the connection interface of the detection equipment, that is: use an appropriate connection cable to connect the industrial control motherboard of the computer to be detected to the detection equipment, so that the industrial control motherboard of the computer to be detected can be powered on normally, and the connection is firm and not loose, so that the detection equipment can receive the operation signal of the industrial control motherboard of the computer to be detected, and finally, confirm in the control software of the detection equipment that the industrial control motherboard of the computer to be detected has been correctly connected and identified, thereby completing the connection of the industrial control motherboard of the computer to be detected to the detection equipment. In other embodiments, other methods can also be used to achieve it, which is not limited here.

[0023] In step 101, images corresponding to each solder joint of the industrial control motherboard of the computer to be inspected when it is running in the inspection device are collected to obtain all solder joint images, and then determine the pixel contribution of the solder joint in each solder joint image.

[0024] In a specific implementation, the image acquisition device in the detection equipment of the computer industrial control motherboard can be used to automatically scan the entire running computer industrial control motherboard to be detected, and collect images at each solder joint one by one, so as to obtain all solder joint images of the computer industrial control motherboard to be detected during operation. The image acquisition device can be, for example, a high-resolution optical microscope, a scanning electron microscope, an automatic optical inspection equipment (Automated Optical Inspection, AOI equipment) and a machine vision system, etc. In other embodiments, other methods can also be used for collection, which is not specifically limited here.

[0025] In some embodiments, reference Figure 2 As shown in FIG. 1 , this figure is an exemplary flow chart of determining pixel contribution in some embodiments of the present application. In this embodiment, determining the pixel contribution of each solder joint in the solder joint image can be implemented by the following steps: First, in step 1011, a welding spot image is selected; Next, in step 1012, each pixel in the weld spot image is divided into defective pixel points and normal pixel points based on a preset determination pixel value; Then, in step 1013, the pixel contribution of the solder joint in the solder joint image is determined based on all defective pixels and all normal pixels; Finally, in step 1014, the above steps are repeated to obtain pixel contributions of the welding points in the remaining welding point images.

[0026] It should be noted that the defective pixel points in the present application represent the pixel points corresponding to defects (for example, false solder joints, cold solder joints, oxidation defects, etc.) at the solder joints of the industrial control motherboard of the computer to be tested. In addition, the normal pixel points represent the pixel points that appear at the solder joints of the industrial control motherboard of the computer to be tested under normal circumstances.

[0027] In specific implementation, the judgment pixel value in the present application is a pixel grayscale threshold used to distinguish defective pixels and normal pixels in a solder joint image. The division of each pixel in the solder joint image into defective pixels and normal pixels based on a preset judgment pixel value can be implemented in the following manner, namely: first, the judgment pixel value can be set by relevant experts based on the pixel value conditions presented by solder joint defects (for example, false solder joints, cold solder joints, oxidation and other defects) in historical computer industrial control motherboards, and then, the pixels in the solder joint image whose grayscale pixel values ​​are less than or equal to the judgment pixel value are all judged as defective pixels, and the pixels in the solder joint image whose grayscale pixel values ​​are greater than the judgment pixel value are all judged as normal pixels. In other embodiments, other methods can also be used for determination, which is not limited here.

[0028] In a specific implementation, the pixel contribution of the solder joint in the solder joint image is determined by all defective pixels and all normal pixels in the following manner, namely: first, the total number of pixels in the solder joint image is obtained, and then the absolute value of the difference between the number of all defective pixels and the number of all normal pixels is multiplied by the number of all defective pixels, and finally, the ratio of the obtained product value to the total number of pixels is used as the pixel contribution of the solder joint in the solder joint image. In other embodiments, other methods may also be used for determination, which is not limited here.

[0029] It should be noted that the pixel contribution of the solder joint in the present application reflects the proportion of defective pixels in the corresponding solder joint image. The higher the pixel contribution of the solder joint, the higher the proportion of defective pixels in the corresponding solder joint image. Conversely, the lower the pixel contribution of the solder joint, the lower the proportion of defective pixels in the corresponding solder joint image.

[0030] In step 102, feature decomposition is performed on the pixels in all solder joint images based on the pixel contribution of each solder joint to obtain a global feature vector of the solder joint on a global defect and a local feature vector of the solder joint on a local defect, and the global feature vector and the local feature vector are fused into a defect description of the solder joint in the industrial control motherboard of the computer to be inspected.

[0031] In some embodiments, the pixel points in all solder joint images are feature decomposed based on the pixel contribution of each solder joint to obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, which can be achieved by the following steps: The global feature space and local feature space of the solder joint defects in the computer industrial control motherboard to be inspected are determined by the pixel points in each solder joint image and the pixel contribution of each solder joint; Decomposing the global feature space to obtain the global feature vector of the solder joints of the industrial control motherboard of the computer to be detected on the global defect; The local feature space is decomposed to obtain the local feature vector of the solder joint on the local defect of the industrial control mainboard of the computer to be detected.

[0032] In specific implementation, the global feature space in the present application represents mapping the correlation between each solder joint image in the industrial control motherboard of the computer to be detected to obtain a high-dimensional space, and the local feature space represents mapping the correlation between the solder joint pixels in the industrial control motherboard of the computer to be detected to obtain a high-dimensional space; the global feature space and local feature space of the solder joint defects in the industrial control motherboard of the computer to be detected are determined by the pixel points in each solder joint image and the pixel contribution of each solder joint. The following method can be used to implement it, namely: the degree of correlation between each two solder joint images can be evaluated by a deep learning algorithm or a machine learning algorithm in combination with the pixel contribution of each solder joint, and then All evaluation values ​​obtained by the kernel method constitute the global feature space of the solder joint defects in the computer industrial control motherboard to be detected; similarly, through a deep learning algorithm or a machine learning algorithm and combined with the judgment result of the pixel point (that is, whether the pixel point is a defective pixel point or a normal pixel point), the degree of correlation between every two pixels in each solder joint image is evaluated, and then all evaluation values ​​obtained by the kernel method constitute the local feature space of the solder joint defects in the computer industrial control motherboard to be detected, wherein the kernel method is, for example: linear kernel, Gaussian kernel, Laplace kernel and Bayesian kernel, etc.; in other embodiments, other methods can also be used for determination, which is not limited here.

[0033] In specific implementation, the global feature space is decomposed to obtain the global feature vector of the solder joints of the computer industrial control motherboard to be detected on the global defects. This can be achieved in the following way, namely: multiple global principal components can be decomposed from the global feature space by principal component analysis (PCA), and one global principal component corresponds to a global defect of the solder joints of the computer industrial control motherboard to be detected. The global defects of the solder joints may include: uneven distribution of solder joints, poor solder joints on the entire board, overall oxidation and corrosion, unqualified overall solder joint size, solder joint offset and solder joint coplanarity problems, etc., and then all the global principal components decomposed in the global feature space are combined to form the global feature vector of the solder joints of the computer industrial control motherboard to be detected on the global defects; the local feature space is decomposed to obtain the local feature vector of the solder joints of the computer industrial control motherboard to be detected on the local defects. The following method is used to achieve this, namely, multiple local principal components can be decomposed from the local feature space by the principal component analysis method, and one local principal component corresponds to a local defect of a solder joint of the industrial control mainboard of the computer to be detected. The local defects of the solder joints may include: cold solder joints, solder joint cracks, solder joint holes, too many or too few solder joints, cold solder joints, solder joint contamination and solder joint bridging, etc., and then all the local principal components decomposed in the local feature space are combined to form a local feature vector of the solder joint of the industrial control mainboard of the computer to be detected on the local defect, that is, feature decomposition of pixels in all solder joint images is achieved; in other embodiments, other methods may also be used for determination, which is not limited here.

[0034] It should be noted that the global feature vector in the present application is an array of feature vectors representing the overall solder point defects of the computer industrial control motherboard to be detected, and the global feature vector reflects the defect situation of the solder points of the computer industrial control motherboard to be detected on the global defects. In addition, the local feature vector is an array of feature vectors representing the local solder point defects of the computer industrial control motherboard to be detected, and the local feature vector reflects the defect situation of the solder points of the computer industrial control motherboard to be detected on the local defects.

[0035] It should be noted that the feature decomposition in the present application refers to the process of mapping the correlation of each solder joint image in the computer industrial control motherboard to be tested to obtain a high-dimensional space (i.e., the global feature space) and mapping the correlation between the solder joint pixels in the computer industrial control motherboard to be tested to obtain a high-dimensional space (i.e., the local feature space), and then decomposing from the global feature space to obtain the global feature vector of the solder joints of the computer industrial control motherboard to be tested on the global defects, and decomposing from the local feature space to obtain the local feature vector of the solder joints of the computer industrial control motherboard to be tested on the local defects; through the principal components decomposed from the global feature space and the local feature space, the global characteristics and changes of the solder joint defects as well as the local characteristics and changes can be revealed in detail, so as to classify and identify different types of global defects and local defects in the computer industrial control motherboard, and at the same time, it can filter noise, reduce dimensionality and facilitate visualization, so as to more accurately detect solder joint defects and improve the quality control effect of the computer industrial control motherboard.

[0036] In some embodiments, reference Figure 3 As shown, this figure is an exemplary flow chart for determining the defect description amount in some embodiments of the present application. In this embodiment, the global feature vector and the local feature vector are merged into the defect description amount of the solder joint in the industrial control motherboard of the computer to be detected, which can be achieved by the following steps: First, in step 1021, the global fusion impact factor of the global feature vector is determined; Next, in step 1022, a local fusion influence factor of the local feature vector is determined; Finally, in step 1023, the global feature vector and the local feature vector are linearly fused by using the global fusion influence factor and the local fusion influence factor to obtain a defect description of the solder joints in the industrial control motherboard of the computer to be inspected.

[0037] In specific implementation, the global fusion influence factor in the present application reflects the degree of influence of external environmental factors on the global defects of solder joints in the industrial control motherboard of the computer to be detected when being integrated. The larger the global fusion influence factor, the greater the degree of influence of external environmental factors on the global defects of solder joints in the industrial control motherboard of the computer to be detected. On the contrary, the smaller the global fusion influence factor, the smaller the degree of influence of external environmental factors on the global defects of solder joints in the industrial control motherboard of the computer to be detected. The global fusion influence factor of the global feature vector can be determined in the following way, namely: the degree of influence of external environmental factors (such as temperature, humidity, vibration, dust and electromagnetic interference, etc.) on the global defects of solder joints in the industrial control motherboard of the computer to be detected can be comprehensively evaluated by combining the evaluation algorithm with historical experimental data, and the obtained evaluation value is used as the global fusion influence factor of the global feature vector. The value of the global fusion influence factor is usually between 0 and 1, wherein the evaluation algorithm is, for example, cross-validation, genetic algorithm, ensemble learning and reinforcement learning, etc., and other methods can also be used for determination in other embodiments, which are not limited here.

[0038] In specific implementation, the local fusion influence factor reflects the influence of internal factors on the local defects of the solder joints in the industrial control mainboard of the computer to be detected when being fused. The larger the local fusion influence factor, the greater the influence of the internal factors on the local defects of the solder joints in the industrial control mainboard of the computer to be detected. On the contrary, the smaller the local fusion influence factor, the smaller the influence of the internal factors on the local defects of the solder joints in the industrial control mainboard of the computer to be detected. The local fusion influence factor of the local feature vector can be determined in the following way, that is, the influence of internal factors (welding temperature, solder composition, welding speed, welding pressure and PCB material of the industrial control mainboard of the computer to be detected) on the local defects of the solder joints in the industrial control mainboard of the computer to be detected can be comprehensively evaluated by combining the evaluation algorithm with historical experimental data, and the obtained evaluation value is used as the local fusion influence factor of the local feature vector. The value of the local fusion influence factor is usually between 0 and 1. The evaluation algorithm is, for example, cross-validation, genetic algorithm, ensemble learning and reinforcement learning, etc. In other embodiments, other methods can also be used for determination, which is not limited here.

[0039] In specific implementation, the global feature vector and the local feature vector are linearly fused through the global fusion influence factor and the local fusion influence factor to obtain the defect description of the solder joints in the industrial control motherboard of the computer to be detected. The following method can be used, namely: first, the global feature vector and the local feature vector can be standardized by the StandardScaler function in the sklearn library in Python to obtain a standardized global feature vector and a standardized local feature vector. Secondly, the global fusion influence factor is multiplied by the standardized global feature vector to obtain an array, and the local fusion influence factor is multiplied by the standardized local feature vector to obtain another array. Then, the two arrays obtained above are added to obtain a group of data. Finally, the average value of the group of data is used as the defect description of the solder joints in the industrial control motherboard of the computer to be detected. In other embodiments, other methods can also be used for determination, which is not limited here.

[0040] It should be noted that the defect description amount in the present application reflects the defect degree of the solder joints in the industrial control motherboard of the computer to be tested. The larger the defect description amount, the higher the defect degree of the solder joints in the industrial control motherboard of the computer to be tested. Conversely, the smaller the defect description amount, the lower the defect degree of the solder joints in the industrial control motherboard of the computer to be tested.

[0041] In step 103, a historical motherboard solder point defect sample set is obtained, and each solder point image is similarly matched with the historical motherboard solder point defect sample set to obtain a defect similarity coefficient between each solder point image and the historical motherboard solder point defect sample set, and the degree of solder point damage of the industrial control motherboard to be tested during operation is determined based on all defect similarity coefficients and the defect description amount.

[0042] It should be noted that the historical motherboard solder point defect sample set is a set composed of defective solder point sample images of historical computer industrial control motherboards. In specific implementation, the defective solder point sample images of historical computer industrial control motherboards can be obtained from the computer industrial control motherboard database, and then all the defective solder point sample images are combined into a historical motherboard solder point defect sample set of the computer industrial control motherboard. In other embodiments, other methods can also be used to obtain it, which is not specifically limited here.

[0043] In some embodiments, similarity matching is performed between each solder joint image and the historical motherboard solder joint defect sample set to obtain a defect similarity coefficient between each solder joint image and the historical motherboard solder joint defect sample set. The following steps can be used to achieve this: Select a solder joint image; Determine a matching value between the solder joint image and each defective solder joint sample image in the historical mainboard solder joint defect sample set; Determine the defect similarity coefficient between the solder joint image and the historical motherboard solder joint defect sample set based on all matching values; Repeat the above steps to obtain the defect similarity coefficients between the remaining solder joint images and the historical motherboard solder joint defect sample set.

[0044] In specific implementation, the matching value reflects the matching degree between the corresponding solder joint image and the defective solder joint sample image in the historical mainboard solder joint defect sample set. The larger the matching value, the greater the matching degree between the corresponding solder joint image and the defective solder joint sample image in the historical mainboard solder joint defect sample set, and the smaller the matching value, the smaller the matching degree between the corresponding solder joint image and the defective solder joint sample image in the historical mainboard solder joint defect sample set. The matching value between the solder joint image and each defective solder joint sample image in the historical mainboard solder joint defect sample set can be achieved in the following manner, namely: selecting the historical mainboard solder joint image and the defective solder joint sample image in the historical mainboard solder joint defect sample set. For a defective solder joint sample image in the point defect sample set, the Euclidean distance between the defective solder joint sample image and the pixel points at each corresponding position in the solder joint image can be calculated by the numpy.linalg.norm() function in Python, and the average value of all the Euclidean distances obtained is used as the matching value between the solder joint image and the defective solder joint sample image. The above steps are repeated to obtain the matching value between the solder joint image and the remaining defective solder joint sample images in the historical motherboard solder joint defect sample set. In other embodiments, other methods can also be used for determination, which is not limited here.

[0045] It should be noted that the defect similarity coefficient in the present application reflects the degree of similarity between the corresponding solder joint image and the historical motherboard solder joint defect sample set. The larger the defect similarity coefficient is, the higher the degree of similarity between the corresponding solder joint image and the historical motherboard solder joint defect sample set is. Conversely, the smaller the defect similarity coefficient is, the smaller the degree of similarity between the corresponding solder joint image and the historical motherboard solder joint defect sample set is. It will not be repeated here. As a preferred embodiment, the defect similarity coefficient between the solder joint image and the historical motherboard solder joint defect sample set is determined by all matching values. The following method can be used to achieve it, namely: the ratio of the largest matching value among all matching values ​​to the average value of all matching values ​​can be used as the defect similarity coefficient between the solder joint image and the historical motherboard solder joint defect sample set. In other embodiments, other methods can also be used for determination, which is not limited here.

[0046] In some embodiments, determining the degree of solder joint damage of the industrial control motherboard of the computer to be inspected during operation by using all defect similarity coefficients and the defect description quantity can be achieved by using the following steps: Determine the comprehensive matching degree between the industrial control motherboard of the computer to be tested and the solder joint defect sample set of the historical motherboard according to all defect similarity coefficients; The degree of damage to the solder joints of the industrial control mainboard of the computer to be inspected during operation is determined by the comprehensive matching degree and the defect description quantity.

[0047] In specific implementation, the comprehensive matching degree reflects the comprehensive matching degree between the industrial control mainboard of the computer to be detected and the solder joint defect sample set of the historical mainboard. The greater the comprehensive matching degree, the greater the comprehensive matching degree between the industrial control mainboard of the computer to be detected and the solder joint defect sample set of the historical mainboard. On the contrary, the smaller the comprehensive matching degree, the smaller the comprehensive matching degree between the industrial control mainboard of the computer to be detected and the solder joint defect sample set of the historical mainboard. The comprehensive matching degree between the industrial control mainboard of the computer to be detected and the solder joint defect sample set of the historical mainboard can be determined according to all defect similarity coefficients. The following method can be used to achieve this, that is, all the defect similarity coefficients can be obtained. The average value of the defect similarity coefficients is used as the comprehensive matching degree between the industrial control mainboard of the computer to be detected and the historical mainboard solder joint defect sample set. In other embodiments, other methods can also be used for determination, which is not limited here. Determining the solder joint damage degree of the industrial control mainboard of the computer to be detected during operation by using the comprehensive matching degree and the defect description amount can be achieved in the following manner, namely: the comprehensive matching degree and the defect description amount can be multiplied, and the multiplication result is used as the solder joint damage degree of the industrial control mainboard of the computer to be detected during operation. In other embodiments, other methods can also be used for determination, which is not limited here.

[0048] It should be noted that the solder point damage degree in the present application reflects the severity of the solder point defects of the industrial control mainboard of the computer to be tested during the operation process. The higher the solder point damage degree, the greater the severity of the solder point defects of the industrial control mainboard of the computer to be tested during the operation process. Conversely, the lower the solder point damage degree, the smaller the severity of the solder point defects of the industrial control mainboard of the computer to be tested during the operation process.

[0049] In step 104, the use quality of the industrial control mainboard of the computer to be inspected is predicted based on the degree of damage of the solder joints, and then the industrial control mainboard of the computer to be inspected is marked as good products according to the prediction result.

[0050] In some embodiments, the following steps are used to predict the use quality of the industrial control motherboard of the computer to be inspected based on the degree of damage of the solder joints, and then to mark the industrial control motherboard of the computer to be inspected as good products according to the prediction results: Preset the good product judgment threshold of the industrial control motherboard of the computer to be tested; When the solder joint damage degree is greater than the good product judgment threshold, it is predicted that the use quality of the industrial control motherboard of the computer to be tested is poor, and the industrial control motherboard of the computer to be tested is marked as a defective product; When the solder joint damage degree is less than or equal to the good product determination threshold, it is predicted that the use quality of the industrial control mainboard of the computer to be tested is good, and the industrial control mainboard of the computer to be tested is marked as a good product.

[0051] It should be noted that the good product judgment threshold is a threshold used to determine whether the detected computer industrial control mainboard is a good product. In specific implementation, the good product judgment threshold can be set by relevant experts using prior experience or historical experimental data. For example, the corresponding minimum solder joint damage degree of all computer industrial control mainboards that have been judged as defective products in the historical experimental data can be used as the good product judgment threshold of the computer industrial control mainboard to be tested, or the average value of all good product judgment thresholds in the historical experimental data can be used as the good product judgment threshold of the computer industrial control mainboard to be tested. In other embodiments, other methods can also be used to set the good product judgment threshold, which is not specifically limited here.

[0052] In addition, in another aspect of the present application, in some embodiments, the present application provides a good product prediction system for a computer industrial control motherboard, referring to Figure 4 , which is a schematic diagram of the structure of a good product prediction system for a computer industrial control motherboard according to some embodiments of the present application. The good product prediction system 400 for a computer industrial control motherboard includes: a collection module 401, a processing module 402 and an execution module 403, which are described as follows: The acquisition module 401 in the present application is mainly used to acquire images corresponding to each solder joint when the industrial control motherboard of the computer to be inspected is running in the inspection device, obtain all solder joint images, and then determine the pixel contribution of each solder joint in each solder joint image; Processing module 402, in the present application, the processing module 402 is mainly used to perform feature decomposition on the pixels in all solder joint images based on the pixel contribution of each solder joint, obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, and fuse the global feature vector and the local feature vector into a defect description of the solder joint in the industrial control motherboard of the computer to be detected; The processing module 402 in the present application is also used to obtain a historical motherboard solder joint defect sample set, perform similarity matching between each solder joint image and the historical motherboard solder joint defect sample set, obtain a defect similarity coefficient between each solder joint image and the historical motherboard solder joint defect sample set, and determine the solder joint damage degree of the industrial control motherboard of the computer to be tested during operation based on all defect similarity coefficients and the defect description amount; Execution module 403, in the present application, execution module 403 is mainly used to predict the use quality of the industrial control motherboard of the computer to be tested based on the degree of damage of the solder joints, and then mark the industrial control motherboard of the computer to be tested as good products according to the prediction results.

[0053] The above describes in detail the examples of the good product prediction method and system for computer industrial control motherboards provided by the embodiments of the present application. It can be understood that in order to realize the above functions, the corresponding device includes hardware structures and / or software modules corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present application.

[0054] In some embodiments, the present application also provides a computer device, which includes a memory and a processor, the memory is used to store computer programs, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-mentioned good product prediction method for computer industrial control motherboards.

[0055] In some embodiments, reference Figure 5 , the dotted line in the figure indicates that the unit or the module is optional, and the figure is a schematic diagram of the structure of a computer device that implements the method for predicting good products for a computer industrial control motherboard of the present application. The method for predicting good products for a computer industrial control motherboard in the above embodiment can be Figure 5 The computer device 500 shown in the figure is implemented, and the computer device 500 includes at least one processor 501, a memory 502 and at least one communication unit 505. The computer device 500 can be a terminal device, a server or a chip.

[0056] The processor 501 may be a general-purpose processor or a special-purpose processor. For example, the processor 501 may be a central processing unit (CPU), which may be used to control the computer device 500, execute software programs, and process data of the software programs. The computer device 500 may also include a communication unit 505 to implement signal input (reception) and output (transmission).

[0057] For example, the computer device 500 may be a chip, the communication unit 505 may be an input and / or output circuit of the chip, or the communication unit 505 may be a communication interface of the chip, and the chip may be a component of a terminal device, a network device, or other devices.

[0058] For another example, the computer device 500 may be a terminal device or a server, and the communication unit 505 may be a transceiver of the terminal device or the server, or the communication unit 505 may be a transceiver circuit of the terminal device or the server.

[0059] The computer device 500 may include one or more memories 502, on which a program 504 is stored. The program 504 can be executed by the processor 501 to generate instructions 503, so that the processor 501 performs the method described in the above method embodiment according to the instructions 503. Optionally, data (such as a target audit model) can also be stored in the memory 502. Optionally, the processor 501 can also read the data stored in the memory 502, and the data can be stored at the same storage address as the program 504, or the data can be stored at a different storage address from the program 504.

[0060] The processor 501 and the memory 502 may be provided separately or integrated together, for example, integrated on a system on chip (SOC) of the terminal device.

[0061] It should be understood that each step of the above method embodiment can be completed by a hardware-based logic circuit or software-based instructions in the processor 501. The processor 501 can be a CPU, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, such as discrete gates, transistor logic devices, or discrete hardware components.

[0062] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0063] For example, in some embodiments, the present application also provides a computer-readable storage medium, in which instructions or codes are stored. When the instructions or codes are run on a computer, the computer implements the above-mentioned good product prediction method for a computer industrial control motherboard when executing.

[0064] In summary, in the good product prediction method and system for computer industrial control motherboards disclosed in the embodiments of the present application, by collecting images corresponding to each solder joint when the computer industrial control motherboard to be tested is running in the detection device, all solder joint images are obtained, and then the pixel contribution of the solder joint in each solder joint image is determined; based on the pixel contribution of each solder joint, the pixel points in all solder joint images are feature decomposed to obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, and the global feature vector and the local feature vector are merged into the defect description of the solder joint in the computer industrial control motherboard to be tested; the historical motherboard solder joints are obtained. Point defect sample set, each solder joint image is similarly matched with the historical mainboard solder joint defect sample set, and the defect similarity coefficient between each solder joint image and the historical mainboard solder joint defect sample set is obtained, and the solder joint damage degree of the computer industrial control mainboard to be tested during operation is determined by all the defect similarity coefficients and the defect description quantity; based on the solder joint damage degree, the use quality of the computer industrial control mainboard to be tested is predicted, and then the computer industrial control mainboard to be tested is marked as good products according to the prediction result; the hidden dangers of computer industrial control mainboard solder joint defects can be pre-detected, thereby increasing the credible range of good product detection of computer industrial control mainboards.

[0065] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0066] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for predicting good products of computer industrial control motherboards, characterized in that: The steps include: Collect images corresponding to each solder joint of the industrial control motherboard of the computer to be tested when it is running in the testing equipment, obtain all solder joint images, and then determine the pixel contribution of the solder joint in each solder joint image; Based on the pixel contribution of each solder joint, the pixel points in all solder joint images are feature decomposed to obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, and the global feature vector and the local feature vector are merged into the defect description of the solder joint in the industrial control motherboard of the computer to be detected; Obtain a historical motherboard solder joint defect sample set, perform similarity matching on each solder joint image with the historical motherboard solder joint defect sample set, obtain a defect similarity coefficient between each solder joint image and the historical motherboard solder joint defect sample set, and determine the solder joint damage degree of the industrial control motherboard of the computer to be tested during operation based on all defect similarity coefficients and the defect description amount; The use quality of the industrial control mainboard of the computer to be inspected is predicted based on the degree of damage of the solder joints, and then the industrial control mainboard of the computer to be inspected is marked as good products according to the prediction result.

2. The method according to claim 1, characterized in that Determining the pixel contribution of each solder joint in the solder joint image specifically includes: Select a solder joint image; Based on a preset determination pixel value, each pixel in the solder joint image is divided into a defective pixel and a normal pixel; Determine the pixel contribution of the solder joint in the solder joint image based on all defective pixels and all normal pixels; Repeat the above steps to obtain the pixel contributions of the solder joints in the remaining solder joint images.

3. The method according to claim 1, characterized in that Based on the pixel contribution of each solder joint, the pixel points in all solder joint images are feature decomposed to obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect. Specifically, the following are included: The global feature space and local feature space of the solder joint defects in the computer industrial control motherboard to be inspected are determined by the pixel points in each solder joint image and the pixel contribution of each solder joint; Decomposing the global feature space to obtain the global feature vector of the solder joints of the industrial control motherboard of the computer to be detected on the global defect; The local feature space is decomposed to obtain the local feature vector of the solder joint on the local defect of the industrial control mainboard of the computer to be detected.

4. The method according to claim 1, characterized in that The defect similarity coefficients of each solder joint image and the historical mainboard solder joint defect sample set are obtained by similarly matching each solder joint image with the historical mainboard solder joint defect sample set. Specifically, the defect similarity coefficients of each solder joint image and the historical mainboard solder joint defect sample set are obtained by: Select a solder joint image; Determine a matching value between the solder joint image and each defective solder joint sample image in the historical mainboard solder joint defect sample set; Determine the defect similarity coefficient between the solder joint image and the historical motherboard solder joint defect sample set based on all matching values; Repeat the above steps to obtain the defect similarity coefficients between the remaining solder joint images and the historical motherboard solder joint defect sample set.

5. The method according to claim 1, characterized in that Determining the degree of solder joint damage of the industrial control motherboard of the computer to be tested during operation by all defect similarity coefficients and the defect description quantity specifically includes: Determine the comprehensive matching degree between the industrial control motherboard of the computer to be tested and the solder joint defect sample set of the historical motherboard according to all defect similarity coefficients; The degree of damage to the solder joints of the industrial control mainboard of the computer to be inspected during operation is determined by the comprehensive matching degree and the defect description quantity.

6. The method according to claim 1, characterized in that Predicting the use quality of the industrial control motherboard of the computer to be inspected based on the degree of damage of the solder joints, and then marking the industrial control motherboard of the computer to be inspected as a good product according to the prediction result specifically includes: Preset the good product judgment threshold of the industrial control motherboard of the computer to be tested; When the solder joint damage degree is greater than the good product judgment threshold, it is predicted that the use quality of the industrial control motherboard of the computer to be tested is poor, and the industrial control motherboard of the computer to be tested is marked as a defective product; When the solder joint damage degree is less than or equal to the good product determination threshold, it is predicted that the use quality of the industrial control mainboard of the computer to be tested is good, and the industrial control mainboard of the computer to be tested is marked as a good product.

7. The method according to claim 1, characterized in that The computer industrial control mainboard is a vehicle-mounted industrial control mainboard.

8. A good product prediction system for computer industrial control motherboard, characterized in that: include: An acquisition module is used to acquire images corresponding to each solder joint when the industrial control motherboard of the computer to be tested is running in the testing device, obtain all solder joint images, and then determine the pixel contribution of the solder joint in each solder joint image; The processing module is further used to perform feature decomposition on the pixels in all solder joint images based on the pixel contribution of each solder joint, obtain the global feature vector of the solder joint on the global defect and the local feature vector of the solder joint on the local defect, and fuse the global feature vector and the local feature vector into a defect description of the solder joint in the industrial control motherboard of the computer to be detected; The processing module is also used to obtain a historical motherboard solder joint defect sample set, perform similarity matching between each solder joint image and the historical motherboard solder joint defect sample set, obtain a defect similarity coefficient between each solder joint image and the historical motherboard solder joint defect sample set, and determine the solder joint damage degree of the industrial control motherboard of the computer to be tested during operation based on all defect similarity coefficients and the defect description amount; The execution module is used to predict the use quality of the industrial control motherboard of the computer to be tested based on the solder joint damage degree, and then mark the industrial control motherboard of the computer to be tested as a good product according to the prediction result.

9. A computer device, characterized in that: The computer device includes a memory and a processor, the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the good product prediction method for a computer industrial control motherboard as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions or codes, and when the instructions or codes are executed on a computer, the computer implements the method for predicting good products for a computer industrial control motherboard as described in any one of claims 1 to 7.

Images