High aspect ratio circuit board and method for vehicle-mounted cameras

Abstract

The present disclosure relates to a circuit board with a high aspect ratio and a method for vehicle-mounted cameras. The method comprises: collecting the appearance image of the circuit board after antioxidant treatment by a camera, and using image recognition and analysis technology based on computer vision to segment the circuit board appearance image based on different scales, extract and enhance the semantic features of local regions, so as to automatically judge whether there are defects in the appearance of the circuit board according to the global similarity topological features composed of the similarity information between the enhanced semantics of the local regions of the circuit board appearance image at different scales. In this way, by segmenting the appearance image of the circuit board at different scales, defects on the circuit board can be captured more carefully, and it can be automatically judged whether there are defects in the circuit board, which reduces the subjectivity and human errors of manual detection and improves the intelligence level of appearance inspection of high aspect ratio circuit boards for vehicle-mounted cameras.

Description

Technical Field

[0001] The present disclosure relates to the technical field of vehicle-mounted cameras, and in particular to a circuit board and method with a high aspect ratio for vehicle-mounted cameras. Background Art

[0002] As an important part of modern automobile safety and information systems, the circuit boards of car cameras must be able to withstand vibration, temperature changes, humidity and other harsh environmental conditions. A high aspect ratio circuit board refers to a circuit board with a large ratio of height (or thickness) to the diameter of the guide hole. This design can increase the stacking density of the circuit board, but it also brings manufacturing challenges.

[0003] After the high aspect ratio circuit board for vehicle-mounted cameras is prepared, it is necessary to conduct an appearance inspection of the circuit board. Appearance inspection is an important step to ensure that the quality of the circuit board meets the standards. In particular, any defects in the circuit board, such as rough hole walls, broken lines, and solder joint defects, may affect the function of the camera.

[0004] Manual visual inspection is the most traditional way to inspect the appearance of high aspect ratio circuit boards for automotive cameras. However, this method relies on the operator's experience and subjective judgment, which may lead to inconsistent standards for defect determination, affecting the accuracy and consistency of the inspection results. In addition, some subtle defects may be difficult to detect with traditional inspection technology. For example, for the detection of subtle defects, traditional technologies such as optical microscopes or simple automatic optical inspection (AOI) may have insufficient resolution to capture subtle problems on ultra-fine pitch or ultra-high density circuit boards. These problems may include tiny open circuits, short circuits, solder joint cracks or extremely small contaminants, thus affecting the accuracy of the circuit board appearance inspection.

[0005] Therefore, a high aspect ratio circuit board for vehicle-mounted cameras is desired. Summary of the invention

[0006] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0007] In a first aspect, the present disclosure provides a method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera, the method comprising:

[0008] Providing a substrate;

[0009] After cleaning and drying the surface of the substrate, a photoresist layer is coated on the surface of the substrate;

[0010] Transferring the circuit pattern to the substrate by photolithography to obtain a bare circuit board;

[0011] Drilling the bare circuit board by laser drilling equipment and / or mechanical drilling equipment to obtain a drilled bare circuit board;

[0012] Placing the drilled bare circuit board in a chemical copper plating solution for chemical copper plating to form a copper layer on the drilled substrate;

[0013] Electroplating copper on the copper layer to increase the thickness of the copper layer to obtain an electroplated circuit board;

[0014] Performing an anti-oxidation treatment on the electroplated circuit board to obtain an anti-oxidation treated circuit board;

[0015] Performing an appearance inspection on the circuit board after the anti-oxidation treatment to obtain an appearance inspection result.

[0016] Optionally, performing an appearance inspection on the circuit board after the antioxidant treatment to obtain an appearance inspection result includes: acquiring an appearance image of the circuit board after the antioxidant treatment collected by a camera; performing image block segmentation on the circuit board appearance image at different scales to obtain a sequence of first-scale circuit board local area images and a sequence of second-scale circuit board local area images; performing local area-based semantic feature extraction on the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images, respectively, to obtain a sequence of first-scale circuit board local area semantic feature vectors and a sequence of second-scale circuit board local area semantic feature vectors; extracting the first-scale circuit board local area semantic feature vector sequence and the second-scale circuit board local area semantic feature vector sequence based on .... The sequence of semantic feature vectors is respectively input into a semantic association enhancement module based on feature energy level measurement association radiation to obtain a sequence of first-scale circuit board local area enhanced semantic feature vectors and a sequence of second-scale circuit board local area enhanced semantic feature vectors; the cosine similarity between any group of first-scale circuit board local area enhanced semantic feature vectors and second-scale circuit board local area enhanced semantic feature vectors in the sequence of first-scale circuit board local area enhanced semantic feature vectors and the sequence of second-scale circuit board local area enhanced semantic feature vectors is calculated to obtain a global similarity topological matrix as a global similarity topological feature; based on the global similarity topological feature, an inspection result is obtained, and the inspection result is used to indicate whether there are defects in the appearance of the circuit board.

[0017] Optionally, performing image block segmentation on the circuit board appearance image at different scales to obtain a sequence of first-scale circuit board local area images and a sequence of second-scale circuit board local area images, includes: performing image block segmentation on the circuit board appearance image at a first scale to obtain a sequence of first-scale circuit board local area images; performing image block segmentation on the circuit board appearance image at a second scale to obtain a sequence of second-scale circuit board local area images, wherein the first scale is different from the second scale.

[0018] Optionally, local area-based semantic feature extraction is performed on the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images to obtain a sequence of first-scale circuit board local area semantic feature vectors and a sequence of second-scale circuit board local area semantic feature vectors, including: inputting the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images into a ViT model to obtain a sequence of the first-scale circuit board local area semantic feature vectors and a sequence of the second-scale circuit board local area semantic feature vectors.

[0019] Optionally, the sequence of the first-scale circuit board local area semantic feature vectors and the sequence of the second-scale circuit board local area semantic feature vectors are respectively input into a semantic association enhancement module based on feature energy level measurement association radiation to obtain a sequence of first-scale circuit board local area enhanced semantic feature vectors and a sequence of second-scale circuit board local area enhanced semantic feature vectors, including: for the sequence of the first-scale circuit board local area semantic feature vectors, calculating the feature energy level coefficients of each first-scale circuit board local area semantic feature vector in the sequence of the first-scale circuit board local area semantic feature vectors to obtain a sequence of first-scale circuit board local area semantic feature energy level coefficients; The sequence is arranged as a first-scale circuit board local area semantic feature energy level coefficient input vector; the first-scale circuit board local area semantic feature energy level coefficient input vector is subjected to one-dimensional convolution coding to obtain a first-scale circuit board local area semantic feature energy level radiation association vector; the first-scale circuit board local area semantic feature energy level radiation association vector is subjected to point convolution coding and probabilistic mapping to obtain a first-scale circuit board local area semantic feature energy level radiation association weight vector; the first-scale circuit board local area semantic feature energy level radiation association weight vector is used as a weight vector, and the sequence of the first-scale circuit board local area semantic feature vectors is subjected to weighted processing and residual processing to obtain a sequence of the first-scale circuit board local area enhanced semantic feature vectors.

[0020] Optionally, for the sequence of the first-scale circuit board local area semantic feature vectors, the characteristic energy level coefficients of the first-scale circuit board local area semantic feature vectors in the sequence of the first-scale circuit board local area semantic feature vectors are calculated to obtain a sequence of the first-scale circuit board local area semantic feature energy level coefficients, including: extracting the maximum value of the first-scale circuit board local area semantic feature vectors to obtain the first-scale circuit board local area semantic feature maximum value; respectively calculating the mean and variance of the first-scale circuit board local area semantic feature vectors to obtain the first-scale circuit board local area semantic feature mean and the first-scale circuit board local area semantic feature variance; adding the first-scale circuit board local area semantic feature variance and the preset hyperparameter to obtain a value obtained by multiplying by four to obtain the first-scale circuit board local area semantic feature energy level coefficients. The first energy level coefficient of the semantic feature of the local area of ​​the circuit board; calculating the square of the difference between the maximum value of the semantic feature of the local area of ​​the circuit board at the first scale and the mean value of the semantic feature of the local area of ​​the circuit board at the first scale to obtain the difference value of the semantic feature of the local area of ​​the circuit board at the first scale; adding the modulated first-scale semantic feature variance of the local area of ​​the circuit board obtained by multiplying the variance of the semantic feature of the local area of ​​the circuit board at the first scale by two, the value obtained by multiplying the preset hyperparameter by two, and the difference value of the semantic feature of the local area of ​​the circuit board at the first scale to obtain the second energy level coefficient of the semantic feature of the local area of ​​the circuit board at the first scale; dividing the first energy level coefficient of the semantic feature of the local area of ​​the circuit board at the first scale by the second energy level coefficient of the semantic feature of the local area of ​​the circuit board at the first scale to obtain the energy level coefficient of the semantic feature of the local area of ​​the circuit board at the first scale.

[0021] Optionally, point convolution encoding and probabilistic mapping are performed on the first-scale circuit board local area semantic feature energy level radiation association vector to obtain the first-scale circuit board local area semantic feature energy level radiation association weight vector, including: point convolution encoding the first-scale circuit board local area semantic feature energy level radiation association vector to obtain the first-scale circuit board local area semantic feature energy level radiation convolution association vector; inputting the first-scale circuit board local area semantic feature energy level radiation convolution association vector into a sigmoid function for probabilistic mapping to obtain the first-scale circuit board local area semantic feature energy level radiation association weight vector.

[0022] Optionally, using the weight vector associated with the energy level radiation of the local area semantic feature of the first-scale circuit board as the weight vector, weighted processing and residual processing are performed on the sequence of the local area semantic feature vectors of the first-scale circuit board to obtain a sequence of enhanced semantic feature vectors of the local area of ​​the first-scale circuit board, including: using the feature values ​​of each position in the weight vector associated with the energy level radiation of the local area semantic feature of the first-scale circuit board as the weight value, performing vector-by-vector positional dot multiplication on the sequence of the local area semantic feature vectors of the first-scale circuit board and adding the sequence of the local area semantic feature vectors of the first-scale circuit board to obtain a sequence of enhanced semantic feature vectors of the local area of ​​the first-scale circuit board.

[0023] Optionally, obtaining the inspection result based on the global similarity topological feature includes: inputting the global similarity topological matrix into a classifier-based appearance inspector to obtain the inspection result.

[0024] In a second aspect, the present disclosure provides a high aspect ratio circuit board for vehicle-mounted cameras, wherein the high aspect ratio circuit board for vehicle-mounted cameras is manufactured by the method for preparing the high aspect ratio circuit board for vehicle-mounted cameras as described above.

[0025] The above technical solution is adopted, by collecting the appearance image of the circuit board after the antioxidant treatment by the camera, and using the image recognition and analysis technology based on computer vision to perform image segmentation based on different scales, extract and enhance the semantic features of the local area of ​​the appearance image of the circuit board, so as to automatically judge whether the appearance of the circuit board has defects according to the global similarity topological features composed of the similarity information between the enhanced semantics of the local area of ​​the appearance image of the circuit board at different scales. In this way, by performing image segmentation of the appearance image of the circuit board at different scales, the defects on the circuit board can be captured more carefully, and the circuit board can be automatically judged whether it has defects, which reduces the subjectivity and human errors of manual detection and improves the intelligence level of appearance inspection of high aspect ratio circuit boards for vehicle-mounted cameras.

[0026] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0028] Figure 1 The present invention is a flow chart showing a method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to an exemplary embodiment.

[0029] Figure 2 It is a flowchart of step S108 of a method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to an exemplary embodiment.

[0030] Figure 3 The present invention is a block diagram of a system for preparing a circuit board with a high aspect ratio for vehicle-mounted cameras according to an exemplary embodiment.

[0031] Figure 4 It is a block diagram of an electronic device according to an exemplary embodiment.

[0032] Figure 5 This is an application scenario diagram of a method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to an exemplary embodiment. DETAILED DESCRIPTION

[0033] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0034] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0035] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0036] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0037] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0038] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0039] The specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0040] Figure 1 FIG. 1 is a flow chart showing a method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0041] Step S101, providing a substrate;

[0042] Step S102, after cleaning and drying the surface of the substrate, a photoresist layer is coated on the surface of the substrate;

[0043] Step S103, transferring the circuit pattern to the substrate through a photolithography process to obtain a bare circuit board;

[0044] Step S104, drilling the bare circuit board by laser drilling equipment and / or mechanical drilling equipment to obtain a drilled bare circuit board;

[0045] Step S105, placing the drilled bare circuit board in a chemical copper plating solution for chemical copper plating to form a copper layer on the drilled substrate;

[0046] Step S106, electroplating copper on the copper layer to increase the thickness of the copper layer to obtain a circuit board after electroplating;

[0047] Step S107, performing an anti-oxidation treatment on the electroplated circuit board to obtain an anti-oxidation treated circuit board;

[0048] Step S108: performing appearance inspection on the circuit board after the anti-oxidation treatment to obtain an appearance inspection result.

[0049] Accordingly, in the above-mentioned appearance inspection of the circuit board after the antioxidant treatment to obtain the appearance inspection result, the appearance image of the circuit board after the antioxidant treatment is collected by a camera, and the image recognition and analysis technology based on computer vision is used to segment the appearance image of the circuit board based on different scales, extract and enhance the semantic features of the local area, so as to automatically judge whether there are defects in the appearance of the circuit board according to the global similarity topological features composed of the similarity information between the enhanced semantics of the local areas of the appearance image of the circuit board at different scales. In this way, by segmenting the appearance image of the circuit board at different scales, the defects on the circuit board, including tiny cracks, gaps or other irregularities, can be captured more carefully. At the same time, the system can automatically judge whether there are defects in the circuit board, which reduces the subjectivity and human errors of manual inspection and improves the intelligence level of appearance inspection of high aspect ratio circuit boards for vehicle-mounted cameras.

[0050] Figure 2 FIG. 1 is a flow chart of step S108 of a method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to an exemplary embodiment. Figure 2 As shown, step S108, performing appearance inspection on the circuit board after the anti-oxidation treatment to obtain an appearance inspection result, includes:

[0051] Step S1081, obtaining an appearance image of the circuit board after the anti-oxidation treatment collected by a camera;

[0052] Step S1082, dividing the circuit board appearance image into image blocks at different scales to obtain a sequence of first-scale circuit board local area images and a sequence of second-scale circuit board local area images;

[0053] Step S1083, performing local area-based semantic feature extraction on the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images respectively to obtain a sequence of first-scale circuit board local area semantic feature vectors and a sequence of second-scale circuit board local area semantic feature vectors;

[0054] Step S1084, respectively inputting the sequence of the first-scale circuit board local area semantic feature vectors and the sequence of the second-scale circuit board local area semantic feature vectors into a semantic association enhancement module based on feature energy level measurement association radiation to obtain a sequence of first-scale circuit board local area enhanced semantic feature vectors and a sequence of second-scale circuit board local area enhanced semantic feature vectors;

[0055] Step S1085, calculating the cosine similarity between any set of the first-scale circuit board local area enhanced semantic feature vectors and the second-scale circuit board local area enhanced semantic feature vectors in the sequence of the first-scale circuit board local area enhanced semantic feature vectors and the sequence of the second-scale circuit board local area enhanced semantic feature vectors to obtain a global similarity topological matrix as a global similarity topological feature;

[0056] Step S1086: Based on the global similarity topological feature, an inspection result is obtained, and the inspection result is used to indicate whether there are defects in the appearance of the circuit board.

[0057] Specifically, in the technical solution of the present application, first, an image of the appearance of a circuit board after antioxidant treatment collected by a camera is obtained. Next, it is considered that the appearance features of the circuit board displayed by local areas of different scales in the circuit board appearance image are different. For example, larger-scale features such as the overall outline of the circuit board, the arrangement of pads, and smaller-scale features such as tiny holes or tiny defects on solder joints, and the circuit board appearance defects contained in these local areas of different scales play an important role in the judgment of circuit board defects. Therefore, in the technical solution of the present application, the circuit board appearance image is segmented into image blocks at a first scale to obtain a sequence of first-scale circuit board local area images, and the circuit board appearance image is segmented into image blocks at a second scale to obtain a sequence of second-scale circuit board local area images.

[0058] In one embodiment of the present disclosure, the circuit board appearance image is segmented into image blocks at different scales to obtain a sequence of first-scale circuit board local area images and a sequence of second-scale circuit board local area images, including: segmenting the circuit board appearance image into image blocks at a first scale to obtain a sequence of first-scale circuit board local area images; segmenting the circuit board appearance image into image blocks at a second scale to obtain a sequence of second-scale circuit board local area images, wherein the first scale is different from the second scale.

[0059] Then, considering that the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images both express the semantic information reflected between the local areas in the local image domain. Based on this, in the technical solution of the present application, the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images are processed using a ViT model with a strong ability to extract image semantics to capture and perceive the semantic association features between the local image blocks to obtain a sequence of the first-scale circuit board local area semantic feature vectors and a sequence of the second-scale circuit board local area semantic feature vectors.

[0060] In one embodiment of the present disclosure, local area-based semantic feature extraction is performed on the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images to obtain a sequence of first-scale circuit board local area semantic feature vectors and a sequence of second-scale circuit board local area semantic feature vectors, including: inputting the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images into a ViT model to obtain a sequence of the first-scale circuit board local area semantic feature vectors and a sequence of the second-scale circuit board local area semantic feature vectors.

[0061] Further, in order to enhance the association between the local semantic features of different scales and the expression of key semantic information of the sequence of the first-scale circuit board local area semantic feature vectors and the sequence of the second-scale circuit board local area semantic feature vectors, in the technical solution of the present application, the sequence of the first-scale circuit board local area semantic feature vectors and the sequence of the second-scale circuit board local area semantic feature vectors are respectively input into the semantic association enhancement module based on the feature energy level measurement association radiation to obtain the sequence of the first-scale circuit board local area enhanced semantic feature vectors and the sequence of the second-scale circuit board local area enhanced semantic feature vectors. In particular, the semantic association enhancement module based on the feature energy level measurement association radiation calculates the feature energy level coefficient of each first-scale circuit board local area enhanced semantic feature vector based on the maximum value, mean and variance of each first-scale circuit board local area enhanced semantic feature vector, so as to quantify the semantic energy distribution of each feature, and represent the intrinsic first-scale circuit board semantic density of each feature and its semantic radiation power to the neighboring feature vectors. Then, each obtained feature energy level coefficient is arranged and processed with one-dimensional convolution and point convolution to capture the semantic dependency association relationship between each feature energy level coefficient of the first scale circuit board local area, and further enhance the feature expression ability to obtain the first scale circuit board local area semantic feature energy level radiation association weight vector. Finally, the association weight vector is used to perform weighted and residual processing on the sequence of the first scale circuit board local area semantic feature vectors to enhance the representation ability of the key semantic information of the first scale circuit board local area and the mutual association between the semantic features of each local area, thereby obtaining the sequence of the first scale circuit board local area enhanced semantic feature vectors. Similarly, the processing process of the sequence of the second scale circuit board local area semantic feature vectors is consistent with the sequence of the first scale circuit board local area semantic feature vectors.

[0062] In one embodiment of the present disclosure, the sequence of the first-scale circuit board local area semantic feature vectors and the sequence of the second-scale circuit board local area semantic feature vectors are respectively input into a semantic association enhancement module based on feature energy level measurement association radiation to obtain a sequence of first-scale circuit board local area enhanced semantic feature vectors and a sequence of second-scale circuit board local area enhanced semantic feature vectors, including: for the sequence of the first-scale circuit board local area semantic feature vectors, calculating the feature energy level coefficients of each first-scale circuit board local area semantic feature vector in the sequence of the first-scale circuit board local area semantic feature vectors to obtain a sequence of first-scale circuit board local area semantic feature energy level coefficients; The sequence of level coefficients is arranged as a first-scale circuit board local area semantic feature energy level coefficient input vector; the first-scale circuit board local area semantic feature energy level coefficient input vector is subjected to one-dimensional convolution coding to obtain a first-scale circuit board local area semantic feature energy level radiation association vector; the first-scale circuit board local area semantic feature energy level radiation association vector is subjected to point convolution coding and probabilistic mapping to obtain a first-scale circuit board local area semantic feature energy level radiation association weight vector; the first-scale circuit board local area semantic feature energy level radiation association weight vector is used as a weight vector, and the sequence of the first-scale circuit board local area semantic feature vectors is subjected to weighted processing and residual processing to obtain a sequence of the first-scale circuit board local area enhanced semantic feature vectors.

[0063] Further, in one embodiment of the present disclosure, for the sequence of the first-scale circuit board local area semantic feature vectors, the characteristic energy level coefficients of the first-scale circuit board local area semantic feature vectors in the sequence of the first-scale circuit board local area semantic feature vectors are calculated to obtain a sequence of the first-scale circuit board local area semantic feature energy level coefficients, including: extracting the maximum value of the first-scale circuit board local area semantic feature vectors to obtain the first-scale circuit board local area semantic feature maximum value; respectively calculating the mean and variance of the first-scale circuit board local area semantic feature vectors to obtain the first-scale circuit board local area semantic feature mean and the first-scale circuit board local area semantic feature variance; adding the first-scale circuit board local area semantic feature variance and the preset hyperparameter to obtain a value obtained by multiplying by four Obtain a first energy level coefficient of a local area semantic feature of a first-scale circuit board; calculate the square of a difference between a maximum value of the local area semantic feature of the first-scale circuit board and a mean value of the local area semantic feature of the first-scale circuit board to obtain a difference value of the local area semantic feature of the first-scale circuit board; add a modulated first-scale circuit board local area semantic feature variance obtained by multiplying the first-scale circuit board local area semantic feature variance by two, a value obtained by multiplying the preset hyperparameter by two, and the first-scale circuit board local area semantic feature difference value to obtain a second energy level coefficient of the local area semantic feature of the first-scale circuit board; divide the first energy level coefficient of the local area semantic feature of the first-scale circuit board by the second energy level coefficient of the local area semantic feature of the first-scale circuit board to obtain an energy level coefficient of the local area semantic feature of the first-scale circuit board.

[0064] Furthermore, in one embodiment of the present disclosure, the first-scale circuit board local area semantic feature energy level radiation association vector is point convolution encoded and probabilistically mapped to obtain the first-scale circuit board local area semantic feature energy level radiation association weight vector, including: point convolution encoding the first-scale circuit board local area semantic feature energy level radiation association vector to obtain the first-scale circuit board local area semantic feature energy level radiation convolution association vector; inputting the first-scale circuit board local area semantic feature energy level radiation convolution association vector into a sigmoid function for probabilistic mapping to obtain the first-scale circuit board local area semantic feature energy level radiation association weight vector.

[0065] Furthermore, in one embodiment of the present disclosure, the weight vector associated with the energy level radiation of the local area semantic feature of the first-scale circuit board is used as the weight vector, and the sequence of the local area semantic feature vectors of the first-scale circuit board is weighted and residual processed to obtain the sequence of the local area enhanced semantic feature vectors of the first-scale circuit board, including: using the feature value of each position in the weight vector associated with the energy level radiation of the local area semantic feature of the first-scale circuit board as the weight value, performing vector-by-vector positional dot multiplication on the sequence of the local area semantic feature vectors of the first-scale circuit board and adding the sequence of the local area semantic feature vectors of the first-scale circuit board to obtain the sequence of the local area enhanced semantic feature vectors of the first-scale circuit board.

[0066] Specifically, the sequence of semantic feature vectors of the local area of ​​the first-scale circuit board is input into the semantic association enhancement module based on the feature energy level measurement association radiation, and is processed by the following semantic association enhancement formula to obtain the sequence of enhanced semantic feature vectors of the local area of ​​the first-scale circuit board; wherein the semantic association enhancement formula is: in, is a sequence of semantic feature vectors of the local area of ​​the first-scale circuit board, and are the first scale circuit board local area semantic feature vectors in the sequence and The first-scale semantic feature vector of the local area of ​​the circuit board, The number of feature vectors in the sequence representing the semantic feature vectors of the local area of ​​the first scale circuit board, and are the first scale circuit board local area semantic feature vectors in the sequence The mean and variance of the semantic feature vectors of the first-scale local area of ​​the circuit board, For the The maximum value of the semantic feature vector of the local area of ​​the first scale circuit board, is the preset hyperparameter, It is The characteristic energy level coefficient of the semantic feature vector of the local area of ​​the first scale circuit board, and They are and The characteristic energy level coefficient of the semantic feature vector of the local area of ​​the first scale circuit board, is the input vector of the semantic feature energy level coefficients of the local area of ​​the first scale circuit board, is a one-dimensional convolutional code, is the scale of the one-dimensional convolution kernel, is the radiation association vector of the semantic feature energy level of the local area of ​​the first scale circuit board, is point convolutional coding, is the function, is the radiation association weight vector of the semantic feature energy level of the local area of ​​the first scale circuit board, It means point multiplication by position. is a sequence of enhanced semantic feature vectors of the local region of the first-scale circuit board. Similarly, the encoding method of the sequence of the local region images of the second-scale circuit board is consistent with the encoding method of the sequence of the local region images of the first-scale circuit board.

[0067] Next, considering that a small defect in the circuit board appearance image and a large defect in the same image may have similar shapes or textures, and at the same time, some components may have similar features or structures at different scales. To better quantify the similarity between the first-scale circuit board local area enhanced semantic features and the second-scale circuit board local area enhanced semantic features, and reveal their similarity or difference in the global image domain, in the technical solution of the present application, the cosine similarity between any set of first-scale circuit board local area enhanced semantic feature vectors and second-scale circuit board local area enhanced semantic feature vectors in the sequence of the first-scale circuit board local area enhanced semantic feature vectors and the sequence of the second-scale circuit board local area enhanced semantic feature vectors is calculated to obtain a global similarity topological matrix, which can comprehensively reflect the similarity or difference relationship between various regions of the circuit board, and provide a basis for subsequent analysis.

[0068] Then, the global similarity topological matrix is ​​input into the classifier-based appearance inspector to obtain the inspection result, and the inspection result is used to indicate whether there is a defect in the appearance of the circuit board. That is, the global similarity topological matrix obtained by similarity calculation using the sequence of the first-scale circuit board local area enhanced semantic feature vectors and the sequence of the second-scale circuit board local area enhanced semantic feature vectors is classified to automatically determine whether there is a defect in the appearance of the circuit board.

[0069] In one embodiment of the present disclosure, obtaining an inspection result based on the global similarity topological feature includes: inputting the global similarity topological matrix into a classifier-based appearance inspector to obtain the inspection result.

[0070] The present application takes into account that the sequence of the first-scale circuit board local area enhanced semantic feature vectors and the sequence of the second-scale circuit board local area enhanced semantic feature vectors respectively express the source semantic context-associated image semantic features of the circuit board appearance image in the local image semantic domain of different scales in the global image semantic domain, based on the associated enhanced features of the energy level measurement of the local image semantic feature domain. In this way, the global similarity topological matrix obtained by calculating the cosine similarity between any set of first-scale circuit board local area enhanced semantic feature vectors and second-scale circuit board local area enhanced semantic feature vectors will also cause insufficient cosine similarity topological association coverage due to the differences in source semantics and feature semantic distribution in the local image semantic domains of each scale, thereby leading to outlier class inference mapping deviation, affecting the accuracy of the inspection results obtained by the global similarity topological matrix through a classifier-based appearance inspector.

[0071] Based on this, in a preferred example, inputting the global similarity topology matrix into a classifier-based appearance inspector to obtain an inspection result includes:

[0072] Calculating the characteristic mean of the global similarity topology matrix, and dividing the characteristic mean by the difference between the maximum eigenvalue and the minimum eigenvalue of the global similarity topology matrix to obtain a global similarity topology characteristic distribution representation value;

[0073] Subtracting the global similarity topological feature distribution representation value from one and dividing it by the global similarity topological feature distribution representation value to obtain a global similarity topological feature distribution modulation value;

[0074] Activating the global similarity topology matrix through a probabilistic function to obtain a probabilistic global similarity topology matrix;

[0075] After performing point subtraction between the probabilistic global similarity topology matrix and the global similarity topology feature distribution modulation value, taking the absolute value and calculating the negative of the logarithmic value with base 2 to obtain a probabilistic global similarity topology feature distribution modulation information vector;

[0076] After dividing the global similarity topology feature distribution representation value by one minus the difference of each eigenvalue of the probabilistic global similarity topology matrix, summing all eigenvalues ​​of the probabilistic global similarity topology matrix and dividing by the scale of the global similarity topology matrix to obtain a probabilistic global similarity topology feature distribution modulation bias value;

[0077] The optimized global similarity topology matrix is ​​obtained by multiplying the probabilistic global similarity topology feature distribution modulation information vector and the probabilistic global similarity topology feature distribution modulation bias value by a weight as a hyperparameter; and

[0078] The optimized global similarity topology matrix is ​​passed through a classifier-based appearance inspector to obtain a control result.

[0079] Wherein, the optimized global similarity topology matrix is ​​expressed as: in, is the probabilistic global similarity topology matrix, is the global similarity topological feature distribution representation value, It is subtracted by position point. represents the logarithmic function value with base 2, It is added by location point. is the weight as a hyperparameter, is the scale of the global similarity topology matrix, i.e., the product of the width and height of the global similarity topology matrix, is each eigenvalue of the probabilistic global similarity topology matrix, is the characteristic mean of the global similarity topology matrix, and are the maximum eigenvalue and the minimum eigenvalue of the global similarity topology matrix, is the optimized global similarity topology matrix.

[0080] That is, in the above preferred example, the eigenvalue-based probability information distribution planning of the global similarity topology matrix is ​​performed by using the Bernoulli probability modulation distribution of the global similarity topology matrix relative to the eigenvalue distribution, and the probability reverse mapping of the probability features of the global similarity topology matrix as a whole is used as an extended coverage of the set mapping space of the global similarity topology matrix to independently understand the interaction path between the intuitive probability information distribution and the abstract probability space mapping of the global similarity topology matrix, so as to improve the accuracy of the inspection results obtained by the classifier-based appearance inspector of the optimized global similarity topology matrix input by avoiding the counterfactual reasoning mapping of the outlier feature distribution of the global similarity topology matrix to the class probability.

[0081] In summary, the above scheme is adopted to collect the appearance image of the circuit board after the antioxidant treatment by the camera, and adopt the image recognition and analysis technology based on computer vision to perform image segmentation based on different scales, extract and enhance the semantic features of local regions on the appearance image of the circuit board, so as to automatically judge whether there are defects in the appearance of the circuit board according to the global similarity topological features composed of the similarity information between the enhanced semantics of the local regions of the appearance image of the circuit board at different scales. In this way, by performing image segmentation of the appearance image of the circuit board at different scales, the defects on the circuit board, including tiny cracks, gaps or other irregularities, can be captured more carefully. At the same time, the system can automatically judge whether there are defects in the circuit board, which reduces the subjectivity and human errors of manual detection and improves the intelligence level of appearance inspection of high aspect ratio circuit boards for vehicle-mounted cameras.

[0082] In one embodiment of the present application, a high aspect ratio circuit board for a vehicle-mounted camera is provided, and the high aspect ratio circuit board for a vehicle-mounted camera is manufactured by the method for preparing a high aspect ratio circuit board for a vehicle-mounted camera as described.

[0083] Figure 3 FIG. 1 is a block diagram of a system for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to an exemplary embodiment. Figure 3 As shown, the preparation system 200 includes:

[0084] The substrate providing module 201 is used to provide a substrate;

[0085] A photoresist layer coating module 202 is used to coat a photoresist layer on the surface of the substrate after cleaning and drying the surface of the substrate;

[0086] A circuit pattern transfer module 203, used to transfer the circuit pattern to the substrate through a photolithography process to obtain a bare circuit board;

[0087] A drilling processing module 204 is used to perform drilling processing on the bare circuit board by using a laser drilling device and / or a mechanical drilling device to obtain a drilled bare circuit board;

[0088] The chemical copper plating module 205 is used to place the drilled bare circuit board in a chemical copper plating solution to perform chemical copper plating to form a copper layer on the drilled substrate;

[0089] The copper electroplating module 206 is used to electroplate copper on the copper layer to increase the thickness of the copper layer to obtain a circuit board after electroplating;

[0090] The anti-oxidation treatment module 207 is used to perform an anti-oxidation treatment on the electroplated circuit board to obtain an anti-oxidation treated circuit board;

[0091] The appearance inspection module 208 is used to perform appearance inspection on the circuit board after the anti-oxidation treatment to obtain an appearance inspection result.

[0092] Reference below Figure 4 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0093] like Figure 4 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0094] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0095] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0096] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0097] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0098] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0099] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0100] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0101] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a module does not limit the module itself in some cases. For example, a test parameter acquisition module may also be described as a "module for acquiring device test parameters corresponding to a target device".

[0102] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0103] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0104] Figure 5 FIG. 1 is an application scenario diagram of a method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to an exemplary embodiment. Figure 5 As shown, in this application scenario, first, an image of the appearance of the circuit board after the anti-oxidation treatment is acquired by a camera (for example, Figure 5 Then, the obtained circuit board appearance image is input to a server (for example, Figure 5 In S) shown in , the server can process the circuit board appearance image based on the preparation of a high aspect ratio circuit board for a vehicle-mounted camera to obtain an inspection result, and the inspection result is used to indicate whether there are defects in the appearance of the circuit board.

[0105] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0106] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0107] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. A method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera, characterized in that: include: Providing a substrate; After cleaning and drying the surface of the substrate, a photoresist layer is coated on the surface of the substrate; Transferring the circuit pattern to the substrate by photolithography to obtain a bare circuit board; Drilling the bare circuit board by laser drilling equipment and / or mechanical drilling equipment to obtain a drilled bare circuit board; Placing the drilled bare circuit board in a chemical copper plating solution for chemical copper plating to form a copper layer on the drilled substrate; Electroplating copper on the copper layer to increase the thickness of the copper layer to obtain an electroplated circuit board; Performing an anti-oxidation treatment on the electroplated circuit board to obtain an anti-oxidation treated circuit board; Performing an appearance inspection on the circuit board after the anti-oxidation treatment to obtain an appearance inspection result; Wherein, performing an appearance inspection on the circuit board after the antioxidant treatment to obtain an appearance inspection result includes: Acquire an appearance image of the circuit board after anti-oxidation treatment captured by a camera; Performing image block segmentation on the circuit board appearance image at different scales to obtain a sequence of first-scale circuit board local area images and a sequence of second-scale circuit board local area images; Performing local area-based semantic feature extraction on the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images respectively to obtain a sequence of first-scale circuit board local area semantic feature vectors and a sequence of second-scale circuit board local area semantic feature vectors; Inputting the sequence of the first-scale circuit board local area semantic feature vectors and the sequence of the second-scale circuit board local area semantic feature vectors into the semantic association enhancement module based on feature energy level measurement association radiation to obtain the sequence of the first-scale circuit board local area enhanced semantic feature vectors and the sequence of the second-scale circuit board local area enhanced semantic feature vectors; Calculating the cosine similarity between any set of the first-scale circuit board local area enhanced semantic feature vectors and the second-scale circuit board local area enhanced semantic feature vectors in the sequence of the first-scale circuit board local area enhanced semantic feature vectors and the sequence of the second-scale circuit board local area enhanced semantic feature vectors to obtain a global similarity topological matrix as a global similarity topological feature; Based on the global similarity topological feature, an inspection result is obtained, and the inspection result is used to indicate whether there is a defect in the appearance of the circuit board.

2. The method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to claim 1, characterized in that: The circuit board appearance image is segmented into image blocks at different scales to obtain a sequence of first-scale circuit board local area images and a sequence of second-scale circuit board local area images, including: Performing image block segmentation on the circuit board appearance image at a first scale to obtain a sequence of local area images of the circuit board at the first scale; The circuit board appearance image is segmented into image blocks at a second scale to obtain a sequence of circuit board local area images at the second scale, wherein the first scale is different from the second scale.

3. The method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to claim 2, characterized in that: The sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images are respectively subjected to local area-based semantic feature extraction to obtain a sequence of first-scale circuit board local area semantic feature vectors and a sequence of second-scale circuit board local area semantic feature vectors, including: inputting the sequence of the first-scale circuit board local area images and the sequence of the second-scale circuit board local area images into a ViT model to obtain a sequence of the first-scale circuit board local area semantic feature vectors and a sequence of the second-scale circuit board local area semantic feature vectors.

4. The method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to claim 3, characterized in that: The sequence of the first-scaled circuit board local area semantic feature vectors and the sequence of the second-scaled circuit board local area semantic feature vectors are respectively input into a semantic association enhancement module based on feature energy level measurement association radiation to obtain a sequence of the first-scaled circuit board local area enhanced semantic feature vectors and a sequence of the second-scaled circuit board local area enhanced semantic feature vectors, including: For the sequence of the first-scaled circuit board local area semantic feature vectors, calculating the feature energy level coefficient of each first-scaled circuit board local area semantic feature vector in the sequence of the first-scaled circuit board local area semantic feature vectors to obtain a sequence of first-scaled circuit board local area semantic feature energy level coefficients; Arranging the sequence of the first-scale circuit board local area semantic feature energy level coefficients into a first-scale circuit board local area semantic feature energy level coefficient input vector; Performing one-dimensional convolution coding on the first-scale circuit board local area semantic feature energy level coefficient input vector to obtain the first-scale circuit board local area semantic feature energy level radiation association vector; Performing point convolution coding and probabilistic mapping on the first-scale circuit board local area semantic feature energy level radiation association vector to obtain the first-scale circuit board local area semantic feature energy level radiation association weight vector; Taking the energy level radiation association weight vector of the semantic feature of the local area of ​​the first-scale circuit board as the weight vector, weighted processing and residual processing are performed on the sequence of the semantic feature vectors of the local area of ​​the first-scale circuit board to obtain the sequence of enhanced semantic feature vectors of the local area of ​​the first-scale circuit board.

5. The method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to claim 4, characterized in that: For the sequence of the first-scaled circuit board local area semantic feature vectors, calculating the feature energy level coefficient of each first-scaled circuit board local area semantic feature vector in the sequence of the first-scaled circuit board local area semantic feature vectors to obtain a sequence of first-scaled circuit board local area semantic feature energy level coefficients, including: Extracting the maximum value of the semantic feature vector of the local area of ​​the first-scale circuit board to obtain the maximum value of the semantic feature of the local area of ​​the first-scale circuit board; Respectively calculating the mean and variance of the semantic feature vector of the local area of ​​the first-scale circuit board to obtain the mean of the semantic feature of the local area of ​​the first-scale circuit board and the variance of the semantic feature of the local area of ​​the first-scale circuit board; The value obtained by adding the variance of the semantic feature of the local area of ​​the first-scale circuit board and the preset hyperparameter is multiplied by four to obtain the first energy level coefficient of the semantic feature of the local area of ​​the first-scale circuit board; Calculating the square of the difference between the maximum value of the semantic feature of the local area of ​​the first-scale circuit board and the mean value of the semantic feature of the local area of ​​the first-scale circuit board to obtain a difference value of the semantic feature of the local area of ​​the first-scale circuit board; The first-scale circuit board local area semantic feature variance obtained by multiplying the first-scale circuit board local area semantic feature variance by two is added to the value obtained by multiplying the preset hyperparameter by two and the first-scale circuit board local area semantic feature difference value to obtain the first-scale circuit board local area semantic feature second energy level coefficient; The first energy level coefficient of the semantic feature of the local area of ​​the first-scale circuit board is divided by the second energy level coefficient of the semantic feature of the local area of ​​the first-scale circuit board to obtain the energy level coefficient of the semantic feature of the local area of ​​the first-scale circuit board.

6. The method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to claim 5, characterized in that: The first-scale circuit board local area semantic feature energy level radiation association vector is subjected to point convolution encoding and probabilistic mapping to obtain the first-scale circuit board local area semantic feature energy level radiation association weight vector, including: Performing point convolution encoding on the first-scale circuit board local area semantic feature energy level radiation association vector to obtain the first-scale circuit board local area semantic feature energy level radiation convolution association vector; The energy level radiation convolution association vector of the semantic feature of the local area of ​​the first-scale circuit board is input into a sigmoid function for probabilistic mapping to obtain a radiation association weight vector of the energy level of the semantic feature of the local area of ​​the first-scale circuit board.

7. The method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to claim 6, characterized in that: The method comprises: taking the weight vector associated with the energy level radiation of the semantic feature of the local area of ​​the first-scale circuit board as the weight vector, performing weighted processing and residual processing on the sequence of the semantic feature vectors of the local area of ​​the first-scale circuit board to obtain the sequence of enhanced semantic feature vectors of the local area of ​​the first-scale circuit board, including: taking the eigenvalues ​​of each position in the weight vector associated with the energy level radiation of the semantic feature of the local area of ​​the first-scale circuit board as the weight value, performing vector-by-vector point multiplication by position on the sequence of the semantic feature vectors of the local area of ​​the first-scale circuit board and adding the sequence of the semantic feature vectors of the local area of ​​the first-scale circuit board to obtain the sequence of enhanced semantic feature vectors of the local area of ​​the first-scale circuit board.

8. The method for preparing a circuit board with a high aspect ratio for a vehicle-mounted camera according to claim 7, characterized in that: Based on the global similarity topological features, obtaining an inspection result includes: inputting the global similarity topological matrix into a classifier-based appearance inspector to obtain the inspection result.

9. A circuit board with a high aspect ratio for vehicle-mounted cameras, characterized in that: The high aspect ratio circuit board for vehicle-mounted cameras is manufactured by the method for manufacturing a high aspect ratio circuit board for vehicle-mounted cameras as described in any one of claims 1-8.

Images