CN116563251A

CN116563251A - Intelligent processing method and system for tie rod ball head for automobile

Info

Publication number: CN116563251A
Application number: CN202310532557.9A
Authority: CN
Inventors: 陈守忠
Original assignee: Yuhuan Ruili Machinery Co ltd
Current assignee: Yuhuan Ruili Machinery Co ltd
Priority date: 2023-05-12
Filing date: 2023-05-12
Publication date: 2023-08-08

Abstract

Discloses an intelligent processing method and system for a tie rod ball head for an automobile. Firstly, performing image preprocessing on an X-Ray detection image of an assembled automobile tie rod ball head, then performing image blocking processing to obtain a sequence of X-Ray detection image blocks, then enabling the sequence of the X-Ray detection image blocks to pass through a ViT model to obtain a plurality of X-Ray detection image block semantic understanding feature vectors, arranging the plurality of X-Ray detection image block semantic understanding feature vectors into a two-dimensional feature matrix, then enabling the two-dimensional feature matrix to pass through a convolutional neural network model containing a bidirectional attention mechanism module to obtain a classification feature matrix, and then enabling an optimized classification feature matrix obtained by optimizing the feature position information expression effect of the classification feature matrix to pass through a classifier to obtain a classification result used for indicating whether the assembled automobile tie rod ball head assembly quality meets a preset standard. In this way, it is possible to determine whether the assembly quality meets a predetermined criterion.

Description

Intelligent processing method and system for tie rod ball head for automobile

Technical Field

The application relates to the field of intelligent machining, and more particularly relates to an intelligent machining method and system for a tie rod ball head for an automobile.

Background

The tie rod ball head for the automobile is an important part for connecting an automobile steering mechanism and a suspension system, and can rotate in different directions so as to adapt to various road conditions in the running process of the automobile.

In order to ensure the safety and stability of the automobile, it is necessary to perform quality detection on the assembled tie rod ball head. The existing method for detecting the quality of the assembled tie rod ball head mainly utilizes manual observation and feeling to judge the rotation flexibility of the tie rod ball head. However, the manual observation and feeling are simple, convenient and quick, but have strong subjectivity, are easily affected by human factors and are difficult to quantify.

Thus, a solution is desired.

Disclosure of Invention

The present application has been made in order to solve the above technical problems. The embodiment of the application provides an intelligent processing method and system for a tie rod ball head for an automobile. Firstly, performing image preprocessing on an X-Ray detection image of an assembled automobile tie rod ball head, then performing image blocking processing to obtain a sequence of X-Ray detection image blocks, then enabling the sequence of the X-Ray detection image blocks to pass through a ViT model to obtain a plurality of X-Ray detection image block semantic understanding feature vectors, arranging the plurality of X-Ray detection image block semantic understanding feature vectors into a two-dimensional feature matrix, then enabling the two-dimensional feature matrix to pass through a convolutional neural network model containing a bidirectional attention mechanism module to obtain a classification feature matrix, and then enabling an optimized classification feature matrix obtained by optimizing the feature position information expression effect of the classification feature matrix to pass through a classifier to obtain a classification result used for indicating whether the assembled automobile tie rod ball head assembly quality meets a preset standard. In this way, it is possible to determine whether the assembly quality meets a predetermined criterion.

According to one aspect of the present application, there is provided an intelligent processing method of a tie rod ball head for an automobile, including: acquiring an X-Ray detection image of the assembled tie rod ball head for the automobile; performing image preprocessing on the X-Ray detection image to obtain a preprocessed X-Ray detection image, wherein the image preprocessing comprises image graying and histogram equalization; performing image blocking processing on the preprocessed X-Ray detection image to obtain a sequence of X-Ray detection image blocks; passing the sequence of X-Ray detection image blocks through a ViT model comprising an embedded layer to obtain a plurality of X-Ray detection image block semantic understanding feature vectors; the semantic understanding feature vectors of the plurality of X-Ray detection image blocks are arranged into a two-dimensional feature matrix, and then a classification feature matrix is obtained through a convolutional neural network model comprising a bidirectional attention mechanism module; optimizing the feature position information expression effect of the classification feature matrix to obtain an optimized classification feature matrix; and the optimized classification feature matrix passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the assembly quality of the assembled tie rod ball head for the automobile meets a preset standard.

In the above-mentioned intelligent processing method of tie rod ball head for automobile, passing the sequence of the X-Ray detection image blocks through ViT model containing embedded layer to obtain a plurality of semantic understanding feature vectors of the X-Ray detection image blocks, including: embedding each X-Ray detection image block in the sequence of X-Ray detection image blocks by using the embedding layer of the ViT model to obtain a sequence of a plurality of X-Ray detection image block embedded vectors; and passing the sequence of the plurality of X-Ray detection image block embedded vectors through the ViT model to obtain the plurality of X-Ray detection image block semantic understanding feature vectors.

In the above-mentioned intelligent processing method of tie rod ball head for automobile, passing the sequence of the embedded vectors of the plurality of X-Ray detection image blocks through the ViT model to obtain semantic understanding feature vectors of the plurality of X-Ray detection image blocks includes: one-dimensional arrangement is carried out on the sequences of the embedding vectors of the plurality of X-Ray detection image blocks so as to obtain global feature vectors; calculating the product between the global feature vector and the transpose vector of each X-Ray detection image block embedding vector in the sequence of the plurality of X-Ray detection image block embedding vectors to obtain a plurality of self-attention correlation matrices; respectively carrying out standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices; obtaining a plurality of probability values by using a Softmax classification function through each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; and weighting each X-Ray detection image block embedded vector in the sequence of the plurality of X-Ray detection image block embedded vectors by taking each probability value in the plurality of probability values as a weight so as to obtain the plurality of X-Ray detection image block semantic understanding feature vectors.

In the above-mentioned intelligent processing method for automotive tie rod ball head, the steps of arranging the semantic understanding feature vectors of the plurality of X-Ray detection image blocks into a two-dimensional feature matrix, and obtaining a classification feature matrix by a convolutional neural network model comprising a bidirectional attention mechanism module include: inputting the two-dimensional feature matrix into the convolutional neural network model to obtain an image block local association feature matrix; and inputting the image block local association feature matrix into the bidirectional attention mechanism module to obtain the classification feature matrix.

In the above-mentioned intelligent processing method of tie rod ball head for automobile, inputting the image block local association feature matrix into the bidirectional attention mechanism module to obtain the classification feature matrix, including: pooling the image block local association feature matrix along the horizontal direction and the vertical direction respectively to obtain a first pooling vector and a second pooling vector; performing association coding on the first pooling vector and the second pooling vector to obtain a bidirectional association matrix; inputting the bidirectional association matrix into a Sigmoid activation function to obtain a bidirectional association weight matrix; and calculating the point-by-point multiplication between the bidirectional association weight matrix and the image block local association feature matrix to obtain the classification feature matrix.

In the above-mentioned intelligent processing method for automotive tie rod ball head, optimizing the feature position information expression effect of the classification feature matrix to obtain an optimized classification feature matrix includes: calculating a location information schema attention response factor for the feature value for each location of the classification feature matrix to obtain a plurality of location information schema attention response factors; and weighting each feature value of the classification feature matrix with the plurality of location information schema attention response factors to obtain the optimized classification feature matrix.

In the above-mentioned intelligent processing method for automotive tie rod ball head, calculating the position information schema attention response factors of the feature values of each position of the classification feature matrix to obtain a plurality of position information schema attention response factors includes: calculating a location information schema attention response factor of the feature value of each location of the classification feature matrix with the following factor calculation formula to obtain a plurality of location information schema attention response factors; wherein, the factor calculation formula is:wherein->Representing a function mapping a two-dimensional real number to a one-dimensional real number, < >>And->The width and the height of the classification feature matrix are respectively +. >For each eigenvalue of the classification eigenvalue matrix +.>Coordinates of (2), and>is the global mean value of all feature values of the classification feature matrix,/for>Represents a logarithmic function with base 2, +.>Is a plurality of location information schema attention response factors.

In the above-mentioned intelligent processing method for automotive tie rod ball head, the classifying result obtained by passing the optimized classifying feature matrix through a classifier is used for indicating whether the assembled quality of the assembled automotive tie rod ball head meets a predetermined standard, and includes: expanding the optimized classification feature matrix into an optimized classification feature vector according to a row vector or a column vector; performing full-connection coding on the optimized classification feature vector by using a plurality of full-connection layers of the classifier to obtain a coding classification feature vector; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

According to another aspect of the present application, there is provided an intelligent machining system for a tie rod ball head for an automobile, comprising: the image acquisition module is used for acquiring an X-Ray detection image of the assembled automobile tie rod ball head; the image preprocessing module is used for carrying out image preprocessing on the X-Ray detection image to obtain a preprocessed X-Ray detection image, wherein the image preprocessing comprises image graying and histogram equalization; the image blocking module is used for carrying out image blocking processing on the preprocessed X-Ray detection image so as to obtain a sequence of X-Ray detection image blocks; the embedded coding module is used for enabling the sequence of the X-Ray detection image blocks to pass through a ViT model containing an embedded layer so as to obtain a plurality of X-Ray detection image block semantic understanding feature vectors; the bidirectional attention coding module is used for arranging the semantic understanding feature vectors of the plurality of X-Ray detection image blocks into a two-dimensional feature matrix and then obtaining a classification feature matrix through a convolutional neural network model comprising a bidirectional attention mechanism module; the optimizing module is used for optimizing the characteristic position information expression effect of the classification characteristic matrix to obtain an optimized classification characteristic matrix; and the classification module is used for enabling the optimized classification feature matrix to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the assembly quality of the assembled tie rod ball head for the automobile meets a preset standard.

In the intelligent processing system of the automobile tie rod ball head, the embedded coding module is used for: embedding each X-Ray detection image block in the sequence of X-Ray detection image blocks by using the embedding layer of the ViT model to obtain a sequence of a plurality of X-Ray detection image block embedded vectors; and passing the sequence of the plurality of X-Ray detection image block embedded vectors through the ViT model to obtain the plurality of X-Ray detection image block semantic understanding feature vectors.

Compared with the prior art, the intelligent processing method and the intelligent processing system for the automobile tie rod ball head are characterized in that firstly, an X-Ray detection image of the assembled automobile tie rod ball head is subjected to image preprocessing and then subjected to image blocking processing to obtain a sequence of X-Ray detection image blocks, then, the sequence of the X-Ray detection image blocks is subjected to ViT model to obtain a plurality of X-Ray detection image block semantic understanding feature vectors, then, the plurality of X-Ray detection image block semantic understanding feature vectors are arranged into a two-dimensional feature matrix and then are subjected to convolutional neural network model comprising a bidirectional attention mechanism module to obtain a classification feature matrix, and then, the optimized classification feature matrix obtained by optimizing the feature position information expression effect of the classification feature matrix is subjected to classifier to obtain a classification result used for indicating whether the assembled automobile tie rod ball head meets a preset standard or not. In this way, it is possible to determine whether the assembly quality meets a predetermined criterion.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. The following drawings are not intended to be drawn to scale, with emphasis instead being placed upon illustrating the principles of the present application.

Fig. 1 is an application scenario diagram of an intelligent processing method of a tie rod ball head for an automobile according to an embodiment of the application.

Fig. 2 is a flowchart of an intelligent processing method of a tie rod ball head for an automobile according to an embodiment of the application.

Fig. 3 is a schematic diagram of an intelligent processing method of a tie rod ball head for an automobile according to an embodiment of the application.

Fig. 4 is a flowchart of substep S140 of the intelligent processing method of the tie rod ball head for the automobile according to the embodiment of the present application.

Fig. 5 is a flowchart of substep S142 of the intelligent machining method of the tie rod ball head for the automobile according to the embodiment of the application.

Fig. 6 is a flowchart of substep S150 of the intelligent processing method of the tie rod ball head for the automobile according to the embodiment of the present application.

Fig. 7 is a flowchart of substep S152 of the intelligent machining method of the tie rod ball head for the automobile according to the embodiment of the application.

Fig. 8 is a flowchart of substep S160 of the intelligent processing method of the tie rod ball head for the automobile according to the embodiment of the application.

Fig. 9 is a flowchart of substep S170 of the intelligent processing method of the tie rod ball head for the automobile according to the embodiment of the application.

Fig. 10 is a block diagram of an intelligent machining system for an automotive tie rod ball according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present application without making any inventive effort, are also within the scope of the present application.

As used in this application and in the claims, the terms "a," "an," "the," and/or "the" are not specific to the singular, but may include the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.

Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server. The modules are merely illustrative, and different aspects of the systems and methods may use different modules.

Flowcharts are used in this application to describe the operations performed by systems according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.

Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.

Aiming at the technical problems, the technical conception of the method is that the X-Ray detection is carried out on the assembled pull rod ball head by utilizing the deep learning and artificial intelligence technology so as to judge whether the assembly quality meets the preset standard or not through an X-Ray image.

Specifically, in the technical scheme of the application, firstly, an X-Ray detection image of an assembled tie rod ball head for an automobile is obtained. Here, by using X-Ray detection, it is possible to penetrate a metal material, showing the internal structure and defects of the assembled tie rod ball for automobiles.

And then, carrying out image preprocessing on the X-Ray detection image to obtain a preprocessed X-Ray detection image, wherein the image preprocessing comprises image graying and histogram equalization. Here, image preprocessing of the X-Ray detection image may improve the quality of the image. Specifically, the image graying is to convert a color image into a gray image, namely, the color value of each pixel is replaced by one gray value, so that the data size of the image can be reduced, and the calculation complexity is reduced; histogram equalization is a method for enhancing image contrast, which can make the gray level distribution of the image more uniform, increase the dynamic range of the image and improve the visual effect of the image.

And then, carrying out image blocking processing on the preprocessed X-Ray detection image to obtain a sequence of X-Ray detection image blocks. Wherein the image blocking process may divide the image into several small areas (image blocks), each of which may be regarded as a separate image. In this way, a more detailed analysis and processing can be performed for each small region.

Since the X-Ray scout image has complex textures, it may be difficult to extract efficient features using conventional convolutional neural network models. In the technical scheme of the application, the sequence of the X-Ray detection image blocks is processed through a ViT model containing an embedded layer to obtain a plurality of X-Ray detection image block semantic understanding feature vectors. Wherein the ViT model is capable of efficiently capturing global semantic information in an image, thereby extracting high-level semantic features. The input to the ViT model is a sequence of image blocks, each of which passes through an embedding layer, converting it into a vector of fixed length. These vectors are then fed into a transform encoder, resulting in a semantically understood feature vector for each image block. These feature vectors may reflect the relationships and dependencies between image blocks.

Further, the plurality of X-Ray detection image block semantic understanding feature vectors are arranged into a two-dimensional feature matrix, and then the two-dimensional feature matrix is obtained through a convolutional neural network model comprising a bidirectional attention mechanism module. In a specific example of the present application, first, the two-dimensional feature matrix is input into the convolutional neural network model to obtain an image block local correlation feature matrix. And then, inputting the image block local association feature matrix into the bidirectional attention mechanism module to obtain the classification feature matrix. Here, the plurality of X-Ray detection image blocks may be arranged in a two-dimensional feature matrix with semantic understanding feature vectors, and spatial position information thereof may be retained. In particular, the convolutional neural network model comprising the bidirectional attention mechanism module can effectively capture the spatial information about the X-Ray detection features in the two-dimensional feature matrix, so that the expression capability of the classification feature matrix is enhanced. In particular, the bi-directional attention mechanism module may leverage contextual information to enhance the detection feature response and suppress the background feature response. The bidirectional attention module respectively calibrates the attention weights of the two-dimensional feature matrix from the horizontal direction and the vertical direction and acquires complex feature relations, so that local feature information can be completely acquired from global features of the space.

And then, the classification feature matrix is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the assembly quality of the assembled tie rod ball head for the automobile meets a preset standard. Namely, the classifier is used for carrying out class boundary division and determination on the high-dimensional data manifold of the classification characteristic matrix so as to obtain a classification label used for indicating whether the assembly quality of the assembled tie rod ball head for the automobile meets the preset standard. Thus, the production and processing process of the tie rod ball head is guided according to the classification result.

In the technical solution of the present application, when the sequence of the X-Ray detection image blocks obtains the plurality of X-Ray detection image block semantic understanding feature vectors through the ViT model including the embedding layer, each X-Ray detection image block semantic understanding feature vector expresses the context-associated image feature semantics of a single X-Ray detection image block relative to other X-Ray detection image blocks, so after the plurality of X-Ray detection image block semantic understanding feature vectors are arranged into a two-dimensional feature matrix, feature values of each position of the two-dimensional feature matrix have corresponding position attributes (i.e., position attributes under the intersection of a semantic distribution dimension and a sample arrangement dimension). And when the two-dimensional feature matrix passes through a convolutional neural network model comprising a bidirectional attention mechanism module, the bidirectional attention mechanism module further strengthens local row space and column space distribution on the rows and columns of the two-dimensional feature matrix, so that the obtained classification feature matrix can be further strengthened, and feature values of all positions have corresponding position attributes.

However, when classifying the classification feature matrix by a classifier, the classification feature matrix needs to be expanded into feature vectors, that is, the classification feature matrix is involvedTo promote the expression effect of the feature position information of each feature value of the classification feature matrix during the permutation transformation, and calculate the position information schema attention response factor of the feature value of each position of the classification feature matrix, specifically expressed as:，/>representing a function mapping two-dimensional real numbers to one-dimensional real numbers, e.g. a representation implemented as a nonlinear activation function activation weighting and biasing,/->And->The width and the height of the classification feature matrix are respectively +.>For each eigenvalue of the classification eigenvalue matrix +.>For example, any vertex of the feature matrix can be used as the origin of coordinates, and +.>Is the global average of all feature values of the classification feature matrix.

Here, the position information schema attention response factor is represented by schema information modeling a relative geometric direction and a relative geometric distance of pixel values with respect to a high-dimensional spatial position of the global feature distribution, and capturing global shape weights of feature manifolds of the high-dimensional feature distribution of the classification feature matrix while achieving position-wise aggregation of feature values with respect to the global feature distribution, so that manifold shapes of the classification feature matrix are highly responsive to shape information of respective sub-manifolds to obtain an arrangement invariance property of the high-dimensional feature manifolds. Therefore, by weighting each characteristic value of the classification characteristic matrix by the position information schema attention response factor, the position information expression effect of each characteristic value of the classification characteristic matrix on the characteristic value of the classification characteristic matrix during arrangement transformation can be improved, and the accuracy of the classification result obtained by the classification characteristic matrix through the classifier is improved.

The technical effects of the application are as follows: 1. the intelligent automobile tie rod ball head assembling quality detection scheme is provided.

2. According to the scheme, by utilizing an X-Ray detection and deep learning technology and analyzing an X-Ray detection image, whether the assembled tie rod ball head has defects or anomalies is accurately and objectively judged, and subjectivity and errors caused by manual observation and feeling are avoided. Meanwhile, the detection efficiency and quality are improved.

Fig. 1 is an application scenario diagram of an intelligent processing method of a tie rod ball head for an automobile according to an embodiment of the application. As shown in fig. 1, in this application scenario, first, an X-Ray detection image (e.g., D shown in fig. 1) of an assembled tie rod ball for an automobile (e.g., N shown in fig. 1) is acquired, and then the X-Ray detection image is input to a server (e.g., S shown in fig. 1) where an intelligent machining algorithm of the tie rod ball for an automobile is deployed, where the server can process the X-Ray detection image using the intelligent machining algorithm of the tie rod ball for an automobile to obtain a classification result for indicating whether the assembly quality of the assembled tie rod ball for an automobile meets a predetermined standard.

Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.

Fig. 2 is a flowchart of an intelligent processing method of a tie rod ball head for an automobile according to an embodiment of the application. As shown in fig. 2, the intelligent processing method for the tie rod ball head for the automobile according to the embodiment of the application includes the following steps: s110, acquiring an X-Ray detection image of the assembled tie rod ball head for the automobile; s120, performing image preprocessing on the X-Ray detection image to obtain a preprocessed X-Ray detection image, wherein the image preprocessing comprises image graying and histogram equalization; s130, performing image blocking processing on the preprocessed X-Ray detection image to obtain a sequence of X-Ray detection image blocks; s140, passing the sequence of the X-Ray detection image blocks through a ViT model containing an embedded layer to obtain a plurality of X-Ray detection image block semantic understanding feature vectors; s150, arranging the semantic understanding feature vectors of the X-Ray detection image blocks into a two-dimensional feature matrix, and then obtaining a classification feature matrix through a convolutional neural network model comprising a bidirectional attention mechanism module; s160, optimizing the feature position information expression effect of the classification feature matrix to obtain an optimized classification feature matrix; and S170, the optimized classification feature matrix is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the assembly quality of the assembled tie rod ball head for the automobile meets a preset standard.

Fig. 3 is a schematic diagram of an intelligent processing method of a tie rod ball head for an automobile according to an embodiment of the application. As shown in fig. 3, in the network architecture, first, an X-Ray detection image of an assembled tie rod ball for an automobile is acquired; then, carrying out image preprocessing on the X-Ray detection image to obtain a preprocessed X-Ray detection image, wherein the image preprocessing comprises image graying and histogram equalization; then, performing image blocking processing on the preprocessed X-Ray detection image to obtain a sequence of X-Ray detection image blocks; then, passing the sequence of X-Ray detection image blocks through a ViT model containing an embedded layer to obtain a plurality of X-Ray detection image block semantic understanding feature vectors; then, arranging the semantic understanding feature vectors of the plurality of X-Ray detection image blocks into a two-dimensional feature matrix, and obtaining a classification feature matrix through a convolutional neural network model comprising a bidirectional attention mechanism module; then, optimizing the feature position information expression effect of the classification feature matrix to obtain an optimized classification feature matrix; and finally, the optimized classification feature matrix passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the assembly quality of the assembled tie rod ball head for the automobile meets a preset standard.

More specifically, in step S110, an X-Ray detection image of the assembled tie rod ball for the automobile is acquired. And the X-Ray detection is utilized, so that the metal material can be penetrated, and the internal structure and defects of the assembled tie rod ball head for the automobile are displayed.

More specifically, in step S120, the X-Ray detection image is subjected to image preprocessing to obtain a preprocessed X-Ray detection image, wherein the image preprocessing includes image graying and histogram equalization. The image preprocessing includes image graying and histogram equalization. Here, image preprocessing of the X-Ray detection image may improve the quality of the image. Specifically, the image graying is to convert a color image into a gray image, namely, the color value of each pixel is replaced by one gray value, so that the data size of the image can be reduced, and the calculation complexity is reduced; histogram equalization is a method for enhancing image contrast, which can make the gray level distribution of the image more uniform, increase the dynamic range of the image and improve the visual effect of the image.

More specifically, in step S130, the preprocessed X-Ray detection image is subjected to an image blocking process to obtain a sequence of X-Ray detection image blocks. The image blocking process may divide the image into several small areas (image blocks), each of which may be regarded as a separate image. In this way, a more detailed analysis and processing can be performed for each small region.

More specifically, in step S140, the sequence of X-Ray detection image blocks is passed through a ViT model containing an embedding layer to obtain a plurality of X-Ray detection image block semantic understanding feature vectors. Since the X-Ray scout image has complex textures, it may be difficult to extract efficient features using conventional convolutional neural network models. In the technical scheme of the application, the sequence of the X-Ray detection image blocks is processed through a ViT model containing an embedded layer to obtain a plurality of X-Ray detection image block semantic understanding feature vectors. Wherein the ViT model is capable of efficiently capturing global semantic information in an image, thereby extracting high-level semantic features. The input to the ViT model is a sequence of image blocks, each of which passes through an embedding layer, converting it into a vector of fixed length. These vectors are then fed into a transform encoder, resulting in a semantically understood feature vector for each image block. These feature vectors may reflect the relationships and dependencies between image blocks.

Accordingly, in one specific example, as shown in fig. 4, passing the sequence of X-Ray detection image blocks through a ViT model containing an embedding layer to obtain a plurality of X-Ray detection image block semantic understanding feature vectors includes: s141, respectively embedding each X-Ray detection image block in the sequence of X-Ray detection image blocks by using the embedding layer of the ViT model to obtain a sequence of a plurality of X-Ray detection image block embedded vectors; and S142, enabling the sequence of the plurality of X-Ray detection image block embedded vectors to pass through the ViT model to obtain the plurality of X-Ray detection image block semantic understanding feature vectors.

Accordingly, in one specific example, as shown in fig. 5, passing the sequence of the plurality of X-Ray detection image block embedding vectors through the ViT model to obtain the plurality of X-Ray detection image block semantic understanding feature vectors includes: s1421, one-dimensionally arranging the sequences of the embedding vectors of the plurality of X-Ray detection image blocks to obtain global feature vectors; s1422, calculating the product between the global feature vector and the transpose vector of each X-Ray detection image block embedding vector in the sequence of the plurality of X-Ray detection image block embedding vectors to obtain a plurality of self-attention correlation matrices; s1423, respectively carrying out standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices; s1424, obtaining a plurality of probability values by using a Softmax classification function for each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; and S1425, weighting each X-Ray detection image block embedded vector in the sequence of the plurality of X-Ray detection image block embedded vectors by taking each probability value in the plurality of probability values as a weight so as to obtain semantic understanding feature vectors of the plurality of X-Ray detection image blocks.

More specifically, in step S150, the plurality of X-Ray detection image blocks are arranged into two-dimensional feature matrices, and then the two-dimensional feature matrices are obtained through a convolutional neural network model including a bidirectional attention mechanism module. The convolutional neural network model comprising the bidirectional attention mechanism module can effectively capture the spatial information about X-Ray detection characteristics in the two-dimensional characteristic matrix, so that the expression capacity of the classification characteristic matrix is enhanced. In particular, the bi-directional attention mechanism module may leverage contextual information to enhance the detection feature response and suppress the background feature response. The bidirectional attention module respectively calibrates the attention weights of the two-dimensional feature matrix from the horizontal direction and the vertical direction and acquires complex feature relations, so that local feature information can be completely acquired from global features of the space.

Accordingly, in one specific example, as shown in fig. 6, after the plurality of X-Ray detection image blocks are arranged into a two-dimensional feature matrix, the two-dimensional feature matrix is obtained through a convolutional neural network model including a bidirectional attention mechanism module, which includes: s151, inputting the two-dimensional feature matrix into the convolutional neural network model to obtain an image block local association feature matrix; and S152, inputting the image block local association feature matrix into the bidirectional attention mechanism module to obtain the classification feature matrix.

Accordingly, in one specific example, as shown in fig. 7, inputting the image block local association feature matrix into the bidirectional attention mechanism module to obtain the classification feature matrix includes: s1521, pooling the image block local association feature matrix along the horizontal direction and the vertical direction respectively to obtain a first pooled vector and a second pooled vector; s1522, performing association coding on the first pooling vector and the second pooling vector to obtain a bidirectional association matrix; s1523, inputting the bidirectional association matrix into a Sigmoid activation function to obtain a bidirectional association weight matrix; and S1524, calculating the point-by-point multiplication between the bidirectional association weight matrix and the image block local association feature matrix to obtain the classification feature matrix.

More specifically, in step S160, the feature position information expression effect of the classification feature matrix is optimized to obtain an optimized classification feature matrix. In the technical solution of the present application, when the sequence of the X-Ray detection image blocks obtains the plurality of X-Ray detection image block semantic understanding feature vectors through the ViT model including the embedding layer, each X-Ray detection image block semantic understanding feature vector expresses the context-associated image feature semantics of a single X-Ray detection image block relative to other X-Ray detection image blocks, so after the plurality of X-Ray detection image block semantic understanding feature vectors are arranged into a two-dimensional feature matrix, feature values of each position of the two-dimensional feature matrix have corresponding position attributes (i.e., position attributes under the intersection of a semantic distribution dimension and a sample arrangement dimension). And when the two-dimensional feature matrix passes through a convolutional neural network model comprising a bidirectional attention mechanism module, the bidirectional attention mechanism module further strengthens local row space and column space distribution on the rows and columns of the two-dimensional feature matrix, so that the obtained classification feature matrix can be further strengthened, and feature values of all positions have corresponding position attributes. However, when the classification feature matrix is classified by a classifier, the classification feature matrix needs to be expanded into feature vectors, that is, a rearrangement transformation based on position attributes involving the feature values of the classification feature matrix, so that in order to promote the feature position information expression effect of the respective feature values of the classification feature matrix at the time of arrangement transformation, a position information pattern attention response factor of the feature values of each position of the classification feature matrix is calculated.

Accordingly, in a specific example, as shown in fig. 8, optimizing the feature location information expression effect of the classification feature matrix to obtain an optimized classification feature matrix includes: s161, calculating a position information schema attention response factor of the feature value of each position of the classification feature matrix to obtain a plurality of position information schema attention response factors; and S162, weighting each characteristic value of the classification characteristic matrix by the plurality of position information schema attention response factors to obtain the optimized classification characteristic matrix.

Accordingly, in one specific example, calculating the location information schema attention response factors for the feature values for each location of the classification feature matrix to obtain a plurality of location information schema attention response factors includes: calculating a location information schema attention response factor of the feature value of each location of the classification feature matrix with the following factor calculation formula to obtain a plurality of location information schema attention response factors; wherein, the factor calculation formula is:wherein->Representing a function mapping a two-dimensional real number to a one-dimensional real number, < >>And->The width and the height of the classification feature matrix are respectively +. >For each eigenvalue of the classification eigenvalue matrix +.>Coordinates of (2), and>is the global mean value of all feature values of the classification feature matrix,/for>Represents a logarithmic function with base 2, +.>Is a plurality of location information schema attention response factors.

More specifically, in step S170, the optimized classification feature matrix is passed through a classifier to obtain a classification result, where the classification result is used to indicate whether the assembly quality of the assembled tie rod ball for the automobile meets a predetermined criterion. And carrying out class boundary division and determination on the high-dimensional manifold of the classification characteristic matrix by using the classifier to obtain a classification label for indicating whether the assembly quality of the assembled tie rod ball head for the automobile meets a preset standard.

That is, in the technical solution of the present application, the label of the classifier includes that the assembled quality of the tie rod ball head for the automobile meets a predetermined standard (first label), and the assembled quality of the tie rod ball head for the automobile does not meet the predetermined standard (second label), where the classifier determines, through a soft maximum function, to which classification label the optimized classification feature matrix belongs. It should be noted that the first tag p1 and the second tag p2 do not include a manually set concept, and in fact, during the training process, the computer model does not have a concept of "whether the assembled quality of the tie rod ball for the assembled automobile meets the predetermined standard", which is simply that there are two kinds of classification tags and the probability that the output feature is under the two kinds of classification tags, that is, the sum of p1 and p2 is one. Therefore, the classification result of whether the assembled tie rod ball head for the automobile meets the preset standard is actually that the classification label is converted into the classification probability distribution conforming to the natural rule, and the physical meaning of the natural probability distribution of the label is actually used instead of the language text meaning of whether the assembled tie rod ball head for the automobile meets the preset standard.

It should be appreciated that the role of the classifier is to learn the classification rules and classifier using a given class, known training data, and then classify (or predict) the unknown data. Logistic regression (logistics), SVM, etc. are commonly used to solve the classification problem, and for multi-classification problems (multi-class classification), logistic regression or SVM can be used as well, but multiple bi-classifications are required to compose multiple classifications, but this is error-prone and inefficient, and the commonly used multi-classification method is the Softmax classification function.

Accordingly, in one specific example, as shown in fig. 9, the optimized classification feature matrix is passed through a classifier to obtain a classification result, where the classification result is used to indicate whether the assembly quality of the assembled tie rod ball head for the automobile meets a predetermined standard, and the method includes: s171, the optimized classification feature matrix is unfolded into an optimized classification feature vector according to a row vector or a column vector; s172, performing full-connection coding on the optimized classification feature vector by using a plurality of full-connection layers of the classifier to obtain a coded classification feature vector; and S173, passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

In summary, according to the intelligent processing method for the automotive tie rod ball head according to the embodiment of the application, firstly, an image preprocessing is performed on an assembled X-Ray detection image of the automotive tie rod ball head, then image blocking processing is performed on the image to obtain a sequence of X-Ray detection image blocks, then, a plurality of X-Ray detection image block semantic understanding feature vectors are obtained through a ViT model by the sequence of X-Ray detection image blocks, then, the plurality of X-Ray detection image block semantic understanding feature vectors are arranged into a two-dimensional feature matrix and then are obtained through a convolutional neural network model comprising a bidirectional attention mechanism module, and then, an optimized classification feature matrix obtained by optimizing the feature position information expression effect of the classification feature matrix is passed through a classifier to obtain a classification result used for indicating whether the assembled automotive tie rod ball head meets a preset standard or not. In this way, it is possible to determine whether the assembly quality meets a predetermined criterion.

Further, according to the technical scheme of the application, a storage medium is further provided, and computer program instructions are stored in the storage medium, and when the computer program instructions are executed by a processor, the processor executes the intelligent processing method of the automotive tie rod ball head.

Fig. 10 is a block diagram of an intelligent machining system 100 for automotive tie rod bulbs according to an embodiment of the present application. As shown in fig. 10, an intelligent processing system 100 for a tie rod ball head for an automobile according to an embodiment of the present application includes: an image acquisition module 110 for acquiring an assembled X-Ray detection image of the tie rod ball head for the automobile; an image preprocessing module 120, configured to perform image preprocessing on the X-Ray detection image to obtain a preprocessed X-Ray detection image, where the image preprocessing includes image graying and histogram equalization; the image blocking module 130 is configured to perform image blocking processing on the preprocessed X-Ray detection image to obtain a sequence of X-Ray detection image blocks; the embedded encoding module 140 is configured to pass the sequence of X-Ray detection image blocks through a ViT model including an embedded layer to obtain a plurality of semantic understanding feature vectors of the X-Ray detection image blocks; the bidirectional attention encoding module 150 is configured to arrange the semantic understanding feature vectors of the plurality of X-Ray detection image blocks into a two-dimensional feature matrix, and then obtain a classification feature matrix through a convolutional neural network model including a bidirectional attention mechanism module; the optimizing module 160 is configured to optimize the feature location information expression effect of the classification feature matrix to obtain an optimized classification feature matrix; and the classification module 170 is configured to pass the optimized classification feature matrix through a classifier to obtain a classification result, where the classification result is used to indicate whether the assembled quality of the assembled tie rod ball head for the automobile meets a predetermined standard.

In one example, in the intelligent processing system 100 for automotive tie rod bulbs described above, the embedded coding module 140 is configured to: embedding each X-Ray detection image block in the sequence of X-Ray detection image blocks by using the embedding layer of the ViT model to obtain a sequence of a plurality of X-Ray detection image block embedded vectors; and passing the sequence of the plurality of X-Ray detection image block embedded vectors through the ViT model to obtain the plurality of X-Ray detection image block semantic understanding feature vectors.

In one example, in the intelligent processing system 100 for automotive tie rod ball head described above, the step of passing the sequence of the plurality of X-Ray detection image block embedded vectors through the ViT model to obtain the plurality of X-Ray detection image block semantic understanding feature vectors includes: one-dimensional arrangement is carried out on the sequences of the embedding vectors of the plurality of X-Ray detection image blocks so as to obtain global feature vectors; calculating the product between the global feature vector and the transpose vector of each X-Ray detection image block embedding vector in the sequence of the plurality of X-Ray detection image block embedding vectors to obtain a plurality of self-attention correlation matrices; respectively carrying out standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices; obtaining a plurality of probability values by using a Softmax classification function through each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; and weighting each X-Ray detection image block embedded vector in the sequence of the plurality of X-Ray detection image block embedded vectors by taking each probability value in the plurality of probability values as a weight so as to obtain the plurality of X-Ray detection image block semantic understanding feature vectors.

In one example, in the intelligent processing system 100 for automotive tie rod bulbs described above, the bi-directional attention encoding module 150 is configured to: inputting the two-dimensional feature matrix into the convolutional neural network model to obtain an image block local association feature matrix; and inputting the image block local association feature matrix into the bidirectional attention mechanism module to obtain the classification feature matrix.

In one example, in the intelligent processing system 100 for automotive tie rod bulbs described above, inputting the image block local correlation feature matrix into the bidirectional attention mechanism module to obtain the classification feature matrix includes: pooling the image block local association feature matrix along the horizontal direction and the vertical direction respectively to obtain a first pooling vector and a second pooling vector; performing association coding on the first pooling vector and the second pooling vector to obtain a bidirectional association matrix; inputting the bidirectional association matrix into a Sigmoid activation function to obtain a bidirectional association weight matrix; and calculating the point-by-point multiplication between the bidirectional association weight matrix and the image block local association feature matrix to obtain the classification feature matrix.

In one example, in the intelligent processing system 100 for automotive tie rod bulbs described above, the optimization module 160 is configured to: calculating a location information schema attention response factor for the feature value for each location of the classification feature matrix to obtain a plurality of location information schema attention response factors; and weighting each eigenvalue of the classification feature matrix with the plurality of location information schema attention response factors to obtain the optimized classification feature matrix.

In one example, in the intelligent processing system 100 for automotive tie rod bulbs described above, calculating the location information schema attention response factors for the eigenvalues of each location of the classification feature matrix to obtain a plurality of location information schema attention response factors includes: calculating a location information schema attention response factor of the feature value of each location of the classification feature matrix with the following factor calculation formula to obtain a plurality of location information schema attention response factors; wherein, the factor calculation formula is:wherein->Representing a function mapping a two-dimensional real number to a one-dimensional real number, < >>And->The width and the height of the classification feature matrix are respectively +. >For each eigenvalue of the classification eigenvalue matrix +.>Coordinates of (2), and>is the global mean value of all feature values of the classification feature matrix,/for>Represents a logarithmic function with base 2, +.>Is a plurality of location information schema attention response factors.

In one example, in the intelligent processing system 100 for automotive tie rod bulbs described above, the classification module 170 is configured to: expanding the optimized classification feature matrix into an optimized classification feature vector according to a row vector or a column vector; performing full-connection coding on the optimized classification feature vector by using a plurality of full-connection layers of the classifier to obtain a coding classification feature vector; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

Here, it will be understood by those skilled in the art that the specific functions and operations of the respective modules in the above-described intelligent processing system 100 for an automotive tie rod ball have been described in detail in the above description of the intelligent processing method for an automotive tie rod ball with reference to fig. 1 to 9, and thus, repetitive descriptions thereof will be omitted.

As described above, the intelligent machining system 100 for an automotive tie rod ball according to the embodiment of the present application may be implemented in various wireless terminals, for example, a server or the like having an intelligent machining algorithm for an automotive tie rod ball. In one example, the intelligent machining system 100 for automotive tie rod bulbs according to embodiments of the present application may be integrated into a wireless terminal as one software module and/or hardware module. For example, the intelligent machining system 100 for automotive tie-rod bulbs may be a software module in the operating system of the wireless terminal, or may be an application developed for the wireless terminal; of course, the intelligent processing system 100 for automotive tie rod bulbs can also be one of a plurality of hardware modules of the wireless terminal.

Alternatively, in another example, the intelligent machining system 100 of the automotive tie rod ball and the wireless terminal may be separate devices, and the intelligent machining system 100 of the automotive tie rod ball may be connected to the wireless terminal through a wired and/or wireless network and transmit interactive information in a agreed data format.

According to another aspect of the present application, there is also provided a non-volatile computer-readable storage medium having stored thereon computer-readable instructions which, when executed by a computer, can perform a method as described above.

Program portions of the technology may be considered to be "products" or "articles of manufacture" in the form of executable code and/or associated data, embodied or carried out by a computer readable medium. A tangible, persistent storage medium may include any memory or storage used by a computer, processor, or similar device or related module. Such as various semiconductor memories, tape drives, disk drives, or the like, capable of providing storage functionality for software.

All or a portion of the software may sometimes communicate over a network, such as the internet or other communication network. Such communication may load software from one computer device or processor to another. For example: a hardware platform loaded from a server or host computer of the video object detection device to a computer environment, or other computer environment implementing the system, or similar functioning system related to providing information needed for object detection. Thus, another medium capable of carrying software elements may also be used as a physical connection between local devices, such as optical, electrical, electromagnetic, etc., propagating through cable, optical cable, air, etc. Physical media used for carrier waves, such as electrical, wireless, or optical, may also be considered to be software-bearing media. Unless limited to a tangible "storage" medium, other terms used herein to refer to a computer or machine "readable medium" mean any medium that participates in the execution of any instructions by a processor.

This application uses specific words to describe embodiments of the application. Reference to "a first/second embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the present application. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various positions in this specification are not necessarily referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the present application may be combined as suitable.

Furthermore, those skilled in the art will appreciate that the various aspects of the invention are illustrated and described in the context of a number of patentable categories or circumstances, including any novel and useful procedures, machines, products, or materials, or any novel and useful modifications thereof. Accordingly, aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.) or by a combination of hardware and software. The above hardware or software may be referred to as a "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present application may take the form of a computer product, comprising computer-readable program code, embodied in one or more computer-readable media.

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

The foregoing is illustrative of the present invention and is not to be construed as limiting thereof. Although a few exemplary embodiments of this invention have been described, those skilled in the art will readily appreciate that many modifications are possible in the exemplary embodiments without materially departing from the novel teachings and advantages of this invention. Accordingly, all such modifications are intended to be included within the scope of this invention as defined in the following claims. It is to be understood that the foregoing is illustrative of the present invention and is not to be construed as limited to the specific embodiments disclosed, and that modifications to the disclosed embodiments, as well as other embodiments, are intended to be included within the scope of the appended claims. The invention is defined by the claims and their equivalents.

Claims

1. An intelligent processing method of a tie rod ball head for an automobile is characterized by comprising the following steps: acquiring an X-Ray detection image of the assembled tie rod ball head for the automobile; performing image preprocessing on the X-Ray detection image to obtain a preprocessed X-Ray detection image, wherein the image preprocessing comprises image graying and histogram equalization; performing image blocking processing on the preprocessed X-Ray detection image to obtain a sequence of X-Ray detection image blocks; passing the sequence of X-Ray detection image blocks through a ViT model comprising an embedded layer to obtain a plurality of X-Ray detection image block semantic understanding feature vectors; the semantic understanding feature vectors of the plurality of X-Ray detection image blocks are arranged into a two-dimensional feature matrix, and then a classification feature matrix is obtained through a convolutional neural network model comprising a bidirectional attention mechanism module; optimizing the feature position information expression effect of the classification feature matrix to obtain an optimized classification feature matrix; and the optimized classification feature matrix passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the assembly quality of the assembled tie rod ball head for the automobile meets a preset standard.

2. The intelligent processing method of a tie rod ball head for an automobile according to claim 1, wherein passing the sequence of X-Ray detection image blocks through a ViT model including an embedded layer to obtain a plurality of X-Ray detection image block semantic understanding feature vectors, comprises: embedding each X-Ray detection image block in the sequence of X-Ray detection image blocks by using the embedding layer of the ViT model to obtain a sequence of a plurality of X-Ray detection image block embedded vectors; and passing the sequence of the plurality of X-Ray detection image block embedded vectors through the ViT model to obtain the plurality of X-Ray detection image block semantic understanding feature vectors.

3. The intelligent processing method of the tie rod ball head for the automobile according to claim 2, wherein the step of passing the sequence of the plurality of X-Ray detection image block embedded vectors through the ViT model to obtain the plurality of X-Ray detection image block semantic understanding feature vectors comprises the steps of: one-dimensional arrangement is carried out on the sequences of the embedding vectors of the plurality of X-Ray detection image blocks so as to obtain global feature vectors; calculating the product between the global feature vector and the transpose vector of each X-Ray detection image block embedding vector in the sequence of the plurality of X-Ray detection image block embedding vectors to obtain a plurality of self-attention correlation matrices; respectively carrying out standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices; obtaining a plurality of probability values by using a Softmax classification function through each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; and weighting each X-Ray detection image block embedded vector in the sequence of the plurality of X-Ray detection image block embedded vectors by taking each probability value in the plurality of probability values as a weight so as to obtain the plurality of X-Ray detection image block semantic understanding feature vectors.

4. The intelligent processing method for a tie rod ball head for an automobile according to claim 3, wherein the steps of arranging the plurality of semantic understanding feature vectors of the X-Ray detection image blocks into a two-dimensional feature matrix and obtaining a classification feature matrix through a convolutional neural network model comprising a bidirectional attention mechanism module comprise: inputting the two-dimensional feature matrix into the convolutional neural network model to obtain an image block local association feature matrix; and inputting the image block local association feature matrix into the bidirectional attention mechanism module to obtain the classification feature matrix.

5. The intelligent processing method of a tie rod ball for an automobile according to claim 4, wherein inputting the image block local correlation feature matrix into the bidirectional attention mechanism module to obtain the classification feature matrix comprises: pooling the image block local association feature matrix along the horizontal direction and the vertical direction respectively to obtain a first pooling vector and a second pooling vector; performing association coding on the first pooling vector and the second pooling vector to obtain a bidirectional association matrix; inputting the bidirectional association matrix into a Sigmoid activation function to obtain a bidirectional association weight matrix; and calculating the point-by-point multiplication between the bidirectional association weight matrix and the image block local association feature matrix to obtain the classification feature matrix.

6. The intelligent processing method for automotive tie rod bulbs according to claim 5, wherein optimizing the feature position information expression effect of the classification feature matrix to obtain an optimized classification feature matrix comprises: calculating a location information schema attention response factor for the feature value for each location of the classification feature matrix to obtain a plurality of location information schema attention response factors; and weighting each feature value of the classification feature matrix with the plurality of location information schema attention response factors to obtain the optimized classification feature matrix.

7. The intelligent processing method of a tie rod ball for a vehicle according to claim 6, wherein calculating a position information pattern attention response factor of the feature value of each position of the classification feature matrix to obtain a plurality of position information pattern attention response factors, comprises: calculated by the following factorsCalculating a position information schema attention response factor of the feature value of each position of the classification feature matrix by a formula to obtain a plurality of position information schema attention response factors; wherein, the factor calculation formula is:wherein->Representing a function mapping a two-dimensional real number to a one-dimensional real number, < > >And->The width and height of the classification feature matrix,for each eigenvalue of the classification eigenvalue matrix +.>Coordinates of (2), and>is the global mean value of all feature values of the classification feature matrix,/for>Represents a logarithmic function with base 2, +.>Is a plurality of location information schema attention response factors.

8. The intelligent processing method of an automotive tie rod ball according to claim 7, wherein the step of passing the optimized classification feature matrix through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the assembly quality of the assembled automotive tie rod ball meets a predetermined standard, and the method comprises the steps of: expanding the optimized classification feature matrix into an optimized classification feature vector according to a row vector or a column vector; performing full-connection coding on the optimized classification feature vector by using a plurality of full-connection layers of the classifier to obtain a coding classification feature vector; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

9. An intelligent processing system for a tie rod ball head for an automobile, comprising: the image acquisition module is used for acquiring an X-Ray detection image of the assembled automobile tie rod ball head; the image preprocessing module is used for carrying out image preprocessing on the X-Ray detection image to obtain a preprocessed X-Ray detection image, wherein the image preprocessing comprises image graying and histogram equalization; the image blocking module is used for carrying out image blocking processing on the preprocessed X-Ray detection image so as to obtain a sequence of X-Ray detection image blocks; the embedded coding module is used for enabling the sequence of the X-Ray detection image blocks to pass through a ViT model containing an embedded layer so as to obtain a plurality of X-Ray detection image block semantic understanding feature vectors; the bidirectional attention coding module is used for arranging the semantic understanding feature vectors of the plurality of X-Ray detection image blocks into a two-dimensional feature matrix and then obtaining a classification feature matrix through a convolutional neural network model comprising a bidirectional attention mechanism module; the optimizing module is used for optimizing the characteristic position information expression effect of the classification characteristic matrix to obtain an optimized classification characteristic matrix; and the classification module is used for enabling the optimized classification feature matrix to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the assembly quality of the assembled tie rod ball head for the automobile meets a preset standard.

10. The intelligent machining system for automotive tie rod bulbs of claim 9, wherein the embedded coding module is configured to: embedding each X-Ray detection image block in the sequence of X-Ray detection image blocks by using the embedding layer of the ViT model to obtain a sequence of a plurality of X-Ray detection image block embedded vectors; and passing the sequence of the plurality of X-Ray detection image block embedded vectors through the ViT model to obtain the plurality of X-Ray detection image block semantic understanding feature vectors.