Method capable of realizing light-emitting sign of front steering wheel

Through deep learning technology, using HOG feature and semantic feature extraction, combined with the significant fusion of feature principal components and the energy-intensive interaction matching network of local fine-grained sequences, the automatic quality detection of the steering wheel luminous sign is realized, solving the consistency and accuracy of traditional manual detection The problem of difficulty in ensuring the sensitivity and reliability of detection has improved.

CN118887214AActive Publication Date: 2024-11-01CIXI ZHUOER PLASTIC PROD

Patent Information

Application Number
CN202411364616.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-29
Publication Date
2024-11-01
Estimated Expiration
2044-09-29

AI Technical Summary

Technical Problem

The quality inspection of traditional steering wheel luminous signs relies on manual visual inspection, which has problems that the consistency and accuracy of the inspection are difficult to ensure. Especially under different lighting conditions, subtle defects may be ignored, affecting the quality and service life of the product.

Method used

The camera is used to collect the steering wheel luminous sign image, and the semantic feature extractor based on the hollow convolution neural network is used to combine the feature principal component significant fusion network to generate multi-dimensional detection features and reference features significant fusion representation vectors, using feature-based Decoupled local fine-grained sequence energy-intensive interaction matching network calculation detection - reference semantic matching coefficients, automatically generate quality detection results.

Benefits of technology

Effectively eliminate the influence of light, reduce interference from human factors, ensure the consistency and reliability of the detection results, and realize the intelligence of the quality inspection of the steering wheel luminous sign.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118887214A_ABST
    Figure CN118887214A_ABST
Patent Text Reader

Abstract

The invention relates to a method capable of achieving a front steering wheel light-emitting sign. The method comprises the steps of collecting a detection steering wheel light-emitting sign image of a front steering wheel light-emitting sign through a camera, and obtaining a reference steering wheel light-emitting sign image which is qualified in quality inspection from a database, an image recognition and analysis technology based on deep learning is adopted to detect and refer to HOG feature extraction and semantic feature extraction of a steering wheel light-emitting sign image, and principal component significant fusion is carried out; therefore, a quality detection result is automatically generated according to a comparison between a semantic matching coefficient between the detection and reference multi-dimensional fusion features and a preset threshold value. Therefore, the defects on the sign can be effectively identified by eliminating the influence of light rays, meanwhile, the interference of human factors is reduced, the consistency and reliability of the detection result are ensured, and the intelligence of the quality detection of the front steering wheel luminous sign is realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of luminous steering wheel signs, and in particular to a method for realizing a front luminous steering wheel sign. Background Art

[0002] In the modern automobile manufacturing industry, with the continuous advancement of automobile interior design, the quality requirements for in-vehicle parts are getting higher and higher. As one of the parts that the driver directly touches, the appearance and functionality of the steering wheel directly affect the driving experience. In particular, the steering wheel with luminous logo can provide better visibility and aesthetics at night or in dim conditions, so its manufacturing process and quality inspection are particularly important.

[0003] However, the traditional quality inspection method of steering wheel illuminated signage relies heavily on manual visual inspection. Specifically, since the inspection results are highly dependent on the subjective judgment of the inspectors, it is difficult to ensure the consistency and accuracy of the inspection. In addition, changes in ambient light will also have a significant impact on the inspection results. Different lighting conditions may lead to inconsistent inspection results. In addition, the human eye is also limited in identifying very small defects, such as fine scratches, color differences or uneven printing. These defects may be ignored in the human eye's inspection, but they may affect the quality and service life of the product.

[0004] Therefore, a method for implementing front steering wheel illuminated logo is desired. Summary of the invention

[0005] 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.

[0006] In a first aspect, the present application provides a method for realizing a front steering wheel luminous mark, the method comprising: Paste the diffusion film on the surface of the light guide plate to obtain a uniform light guide structure; The light-guiding uniform structure, the PCBA board and the base are clamped together to obtain a light-emitting component; The light-emitting assembly is clamped with the light-emitting panel to obtain a front steering wheel light-emitting sign; Performing a quality inspection on the front steering wheel luminous sign to obtain a quality inspection result, wherein the quality inspection result is used to indicate whether the quality of the front steering wheel luminous sign meets a predetermined requirement; Wherein, the front steering wheel luminous sign is subjected to quality inspection to obtain a quality inspection result, and the quality inspection result is used to indicate whether the quality of the front steering wheel luminous sign meets the predetermined requirements, including: A detection steering wheel luminous sign image of the front steering wheel luminous sign collected by a camera; Obtain a reference steering wheel illuminated sign image that has passed quality inspection from a database; The detected steering wheel luminous sign image is passed through a HOG feature extractor to obtain a detected steering wheel luminous sign HOG feature vector; The detected steering wheel luminous sign image is passed through a steering wheel luminous sign semantic feature extractor based on a hole convolutional neural network to obtain a detected steering wheel luminous sign semantic feature vector; The HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign are passed through a feature principal component significance fusion network to obtain a steering wheel luminous sign multi-dimensional detection feature significance fusion representation vector; The reference steering wheel luminous sign image is respectively passed through the HOG feature extractor and the steering wheel luminous sign semantic feature extractor based on the hole convolutional neural network, and the obtained reference steering wheel luminous sign HOG feature vector and reference steering wheel luminous sign semantic feature vector are input into the feature principal component significance fusion network to obtain a steering wheel luminous sign multi-dimensional reference feature significance fusion representation vector; The steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector are passed through a local fine-grained sequence energy dense interactive matching network based on feature decoupling to obtain the steering wheel luminous sign detection-reference semantic matching coefficient; Based on detecting the steering wheel illuminated sign and matching the reference semantics coefficient, the quality detection result is obtained.

[0007] Optionally, the HOG feature vector of the steering wheel luminous sign detection and the semantic feature vector of the steering wheel luminous sign detection are passed through a feature principal component significance fusion network to obtain a steering wheel luminous sign multi-dimensional detection feature significance fusion representation vector, including: standardizing the HOG feature vector of the steering wheel luminous sign detection and the semantic feature vector of the steering wheel luminous sign detection to obtain a standardized steering wheel luminous sign detection HOG feature vector and a standardized steering wheel luminous sign detection semantic feature vector; calculating the sample covariance matrix of the standardized steering wheel luminous sign detection HOG feature vector and the standardized steering wheel luminous sign detection semantic feature vector to obtain a steering wheel luminous sign detection HOG sample covariance matrix and a steering wheel luminous sign detection semantic sample covariance matrix; performing matrix decomposition-based feature vector extraction on the HOG sample covariance matrix of the steering wheel luminous sign detection and the semantic sample covariance matrix of the steering wheel luminous sign detection to obtain a steering wheel luminous sign detection HOG principal component feature vector. A set of quantities and a set of semantic principal component feature vectors of steering wheel luminous signs; input the set of HOG principal component feature vectors of steering wheel luminous signs and the set of semantic principal component feature vectors of steering wheel luminous signs into a maximum approximate query matching network to obtain a set of best matching pairs of HOG principal component feature vectors of steering wheel luminous signs and semantic principal component feature vectors of steering wheel luminous signs; input the best matching pairs of each HOG principal component feature vector of steering wheel luminous signs and semantic principal component feature vector of steering wheel luminous signs in the set of best matching pairs of HOG principal component feature vectors of steering wheel luminous signs and semantic principal component feature vectors of steering wheel luminous signs into a semantic fine-grained gated joint module to obtain a set of HOG-semantic principal component fusion feature vectors of steering wheel luminous signs; cascade the set of HOG-semantic principal component fusion feature vectors of steering wheel luminous signs to obtain a significant fusion representation vector of multi-dimensional detection features of steering wheel luminous signs.

[0008] Optionally, the HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign are standardized to obtain a standardized HOG feature vector of the detected steering wheel luminous sign and a standardized semantic feature vector of the detected steering wheel luminous sign, including: respectively calculating the mean and standard deviation of the HOG feature vector of the detected steering wheel luminous sign to obtain the HOG feature mean of the detected steering wheel luminous sign and the HOG feature standard deviation of the detected steering wheel luminous sign; subtracting the HOG feature vector of the detected steering wheel luminous sign from the HOG feature mean of the detected steering wheel luminous sign by position, and calculating the HOG difference vector of the detected steering wheel luminous sign The standardized HOG feature vector of the detected steering wheel luminous sign is obtained by positionally dividing the standard deviation of the HOG feature of the detected steering wheel luminous sign; the mean and standard deviation of the semantic feature vector of the detected steering wheel luminous sign are respectively calculated to obtain the mean of the semantic features of the detected steering wheel luminous sign and the standard deviation of the semantic features of the detected steering wheel luminous sign; after subtracting the semantic feature vector of the detected steering wheel luminous sign from the mean of the semantic features of the detected steering wheel luminous sign by position, the calculated semantic offset vector of the detected steering wheel luminous sign is divided by the standard deviation of the semantic features of the detected steering wheel luminous sign to obtain the standardized semantic feature vector of the detected steering wheel luminous sign.

[0009] Optionally, the sample covariance matrix of the standardized detection steering wheel luminous sign HOG feature vector and the standardized detection steering wheel luminous sign semantic feature vector is calculated to obtain the detection steering wheel luminous sign HOG sample covariance matrix and the detection steering wheel luminous sign semantic sample covariance matrix, including: multiplying the transposed vector of the standardized detection steering wheel luminous sign HOG feature vector with the standardized detection steering wheel luminous sign HOG feature vector, and then dividing the obtained standardized detection steering wheel luminous sign HOG association matrix by the value obtained by subtracting one from the length of the standardized detection steering wheel luminous sign HOG feature vector by position to obtain the detection steering wheel luminous sign HOG sample covariance matrix; multiplying the transposed vector of the standardized detection steering wheel luminous sign semantic feature vector with the standardized detection steering wheel luminous sign semantic feature vector, and then dividing the obtained standardized detection steering wheel luminous sign semantic association matrix by the value obtained by subtracting one from the length of the standardized detection steering wheel luminous sign semantic feature vector by position to obtain the detection steering wheel luminous sign semantic sample covariance matrix.

[0010] Optionally, the set of HOG principal component feature vectors of the detected steering wheel sign and the set of semantic principal component feature vectors of the detected steering wheel sign are input into a maximum approximate query matching network to obtain a set of best matching pairs of the HOG principal component feature vectors of the detected steering wheel sign and the semantic principal component feature vectors of the detected steering wheel sign, including: extracting a predetermined HOG principal component feature vector of the detected steering wheel sign from the set of HOG principal component feature vectors of the detected steering wheel sign; calculating the cosine similarity between the predetermined HOG principal component feature vector of the detected steering wheel sign and each semantic principal component feature vector of the detected steering wheel sign in the set of semantic principal component feature vectors of the detected steering wheel sign to obtain a set of matching query similarities; and taking the semantic principal component feature vector of the detected steering wheel sign corresponding to the largest matching query similarity in the set of matching query similarities and the predetermined HOG principal component feature vector of the detected steering wheel sign as the best matching pair of the predetermined HOG principal component feature vector of the detected steering wheel sign and the semantic principal component feature vector of the detected steering wheel sign.

[0011] Optionally, each best matching pair of the HOG principal component feature vector for detecting the steering wheel luminous sign and the semantic principal component feature vector for detecting the steering wheel luminous sign is input into a semantic fine-grained gated joint module to obtain a set of HOG-semantic principal component fusion feature vectors for detecting the steering wheel luminous sign, including: respectively calculating the position difference, position dot multiplication and position addition between the best matching pairs of the HOG principal component feature vector for detecting the steering wheel luminous sign and the semantic principal component feature vector for detecting the steering wheel luminous sign to obtain the HOG principal component feature vector for detecting the steering wheel luminous sign G-semantic difference vector, HOG-semantic dot product vector of steering wheel luminous sign detection and HOG-semantic sum vector of steering wheel luminous sign detection; cascade the HOG-semantic difference vector of steering wheel luminous sign detection, the HOG-semantic dot product vector of steering wheel luminous sign detection and the HOG-semantic sum vector of steering wheel luminous sign detection and perform one-dimensional convolution encoding to obtain a multi-dimensional fusion vector of the HOG-semantic principal component pattern of steering wheel luminous sign detection; perform local window-based maximum pooling processing on the multi-dimensional fusion vector of the HOG-semantic principal component pattern of steering wheel luminous sign detection to obtain the HOG-semantic principal component fusion feature vector of steering wheel luminous sign detection.

[0012] Optionally, the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector are passed through a local fine-grained sequence energy-intensive interactive matching network based on feature decoupling to obtain the steering wheel luminous sign detection-reference semantic matching coefficient, including: performing feature decoupling on the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector to obtain a sequence of steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and a sequence of steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors; performing feature selection on the sequence of steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and the sequence of steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors to obtain a selected sequence of steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and a selected sequence of steering wheel luminous sign multi-dimensional reference features. A sequence of locally significant fusion representation vectors; calculating the energy synergy response value between any pair of selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors in the sequence of selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and the sequence of selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors to obtain a steering wheel luminous sign multi-dimensional reference-detection feature synergy energy fine-grained global distribution matrix; performing atrous convolution encoding on the steering wheel luminous sign multi-dimensional reference-detection feature synergy energy fine-grained global distribution matrix to obtain a steering wheel luminous sign multi-dimensional reference-detection feature synergy energy fine-grained global distribution feature matrix; inputting the steering wheel luminous sign multi-dimensional reference-detection feature synergy energy fine-grained global distribution feature matrix into a sequence matching degree measurement network to obtain the steering wheel luminous sign detection-reference semantic matching coefficient.

[0013] Optionally, feature selection is performed on the sequence of local significant fusion representation vectors of the multidimensional detection features of the steering wheel luminous sign and the sequence of local significant fusion representation vectors of the multidimensional reference features of the steering wheel luminous sign to obtain a selected sequence of local significant fusion representation vectors of the multidimensional detection features of the steering wheel luminous sign and a selected sequence of local significant fusion representation vectors of the multidimensional reference features of the steering wheel luminous sign, including: using a first importance measurement function to calculate the steering wheel luminous sign multidimensional detection importance factor of each steering wheel luminous sign multidimensional detection feature local significant fusion representation vector in the sequence of the multidimensional detection features of the steering wheel luminous sign; based on a comparison between each of the multidimensional detection importance factors of the steering wheel luminous sign and a first preset threshold, Performing feature selection on the sequence of local significant fusion representation vectors of multidimensional detection features of steering wheel illuminated signs to obtain the selected sequence of local significant fusion representation vectors of multidimensional detection features of steering wheel illuminated signs; using a second importance measurement function to calculate the steering wheel illuminated sign multidimensional reference importance factor of each steering wheel illuminated sign multidimensional reference feature local significant fusion representation vector in the sequence of multidimensional reference feature local significant fusion representation vectors of steering wheel illuminated signs; based on the comparison between each of the multidimensional reference importance factors of the steering wheel illuminated sign and a second preset threshold, performing feature selection on the sequence of local significant fusion representation vectors of multidimensional reference features of steering wheel illuminated signs to obtain the selected sequence of local significant fusion representation vectors of multidimensional reference features of steering wheel illuminated signs.

[0014] Optionally, calculating the energy synergy response value between any pair of selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors in the sequence of the selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and the selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors to obtain the steering wheel luminous sign multi-dimensional reference-detection feature synergy energy fine-grained global distribution matrix, including: calculating the selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vector and the selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vector The multi-dimensional reference feature of the steering wheel luminous sign is locally significantly fused, and the position-by-position division between the representation vectors is performed to obtain the multi-dimensional local area matching feature vector of the selected steering wheel luminous sign; the maximum value of the multi-dimensional local area matching feature vector of the selected steering wheel luminous sign is extracted to obtain the maximum value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign; the mean and variance of the multi-dimensional local area matching feature vector of the selected steering wheel luminous sign are calculated respectively to obtain the mean value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign and the variance of the multi-dimensional local area matching feature of the selected steering wheel luminous sign; the selected steering wheel luminous sign is The value obtained by adding the variance of the multi-dimensional local area matching feature of the sign and the bias term is multiplied by four to obtain the first energy collaborative response value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign; the square of the difference between the maximum value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign and the mean value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign is calculated to obtain the multi-dimensional local area feature difference value of the selected steering wheel luminous sign; the modulated value of the multi-dimensional local area matching feature variance of the selected steering wheel luminous sign obtained by multiplying the variance of the multi-dimensional local area matching feature of the selected steering wheel luminous sign by two is multiplied by the bias term. The value obtained by multiplying the item by two and the difference value of the multi-dimensional local area feature of the selected steering wheel luminous sign are added to obtain the second energy synergistic response value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign; the first energy synergistic response value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign is divided by the second energy synergistic response value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign to obtain the energy synergistic response value between the local significant fusion representation vector of the multi-dimensional detection feature of the selected steering wheel luminous sign and the local significant fusion representation vector of the multi-dimensional reference feature of the selected steering wheel luminous sign.

[0015] Optionally, obtaining the quality detection result based on the steering wheel illuminated sign detection-reference semantic matching coefficient includes: comparing the steering wheel illuminated sign detection-reference semantic matching coefficient with a predetermined threshold to obtain the quality detection result.

[0016] The above technical solution is adopted to collect the detection steering wheel luminous sign image of the front steering wheel luminous sign through the camera, and obtain the reference steering wheel luminous sign image that has passed the quality inspection from the database, and use the image recognition and analysis technology based on deep learning to extract the HOG features and semantic features of the detection and reference steering wheel luminous sign images, and perform the principal component significant fusion, so as to automatically generate the quality inspection result according to the comparison between the semantic matching coefficient between the detection and reference multi-dimensional fusion features and the predetermined threshold. In this way, the influence of light can be eliminated to effectively identify the defects on the sign, while reducing the interference of human factors, ensuring the consistency and reliability of the inspection results, and realizing the intelligent quality inspection of the front steering wheel luminous sign.

[0017] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other features, advantages and aspects of the embodiments of the present application 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: Figure 1 The present invention is a flow chart showing a method for realizing a front steering wheel luminous logo according to an exemplary embodiment.

[0019] Figure 2 is based on Figure 1 The illustrated embodiment shows a flow chart of step S104 of a method for implementing a front steering wheel luminous mark.

[0020] Figure 3 The invention is a block diagram of a system for realizing a front steering wheel luminous logo according to an exemplary embodiment.

[0021] Figure 4 The figure is a schematic diagram of the structure of a steering wheel luminous sign according to an exemplary embodiment.

[0022] Figure 5 The figure is a schematic diagram of a luminous sign product according to an exemplary embodiment.

[0023] Figure 6 is a schematic diagram of a light emitting panel structure according to an exemplary embodiment.

[0024] Figure 7 is a schematic diagram of a light guide plate structure according to an exemplary embodiment.

[0025] Figure 8 is a schematic diagram of a diffusion membrane structure according to an exemplary embodiment.

[0026] Fig. 9 FIG. 1 is a schematic diagram of a diffusion membrane according to an exemplary embodiment.

[0027] Fig.10 FIG. 1 is a schematic diagram of a PCBA structure according to an exemplary embodiment.

[0028] Fig.11 The figure is a schematic diagram of a whiteboard according to an exemplary embodiment.

[0029] Fig.12 is a circuit design schematic diagram according to an exemplary embodiment.

[0030] Fig.13 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0031] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0032] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0033] 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.

[0034] It should be noted that the concepts such as "first" and "second" mentioned in this application 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.

[0035] It should be noted that the modifications of "one" and "plurality" mentioned in the present application 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".

[0036] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0037] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0038] Figure 1 The present invention is a flow chart showing a method for realizing a front steering wheel luminous logo according to an exemplary embodiment. Figure 2 is based on Figure 1 The embodiment shown is a flow chart of step S104 of a method for implementing a front steering wheel luminous mark. Figure 1 and Figure 2 As shown, the method includes: Step S101, pasting a diffusion film on the surface of a light guide plate to obtain a uniform light guide structure; Step S102, the light-guiding uniform structure, the PCBA board and the base are connected to obtain a light-emitting component; Step S103, clamping the light-emitting component and the light-emitting panel to obtain a front steering wheel light-emitting sign; Step S104, performing a quality inspection on the front steering wheel luminous sign to obtain a quality inspection result, wherein the quality inspection result is used to indicate whether the quality of the front steering wheel luminous sign meets a predetermined requirement; Wherein, step S104, performing quality inspection on the front steering wheel luminous sign to obtain a quality inspection result, wherein the quality inspection result is used to indicate whether the quality of the front steering wheel luminous sign meets the predetermined requirements, includes: Step S1041, detecting the luminous steering wheel sign image of the front luminous steering wheel sign collected by a camera; Step S1042, obtaining a reference steering wheel luminous sign image that has passed quality inspection from a database; Step S1043, passing the detected steering wheel luminous sign image through a HOG feature extractor to obtain a detected steering wheel luminous sign HOG feature vector; Step S1044, passing the detected steering wheel luminous sign image through a steering wheel luminous sign semantic feature extractor based on a dilated convolutional neural network to obtain a detected steering wheel luminous sign semantic feature vector; Step S1045, the HOG feature vector of the detected luminous steering wheel sign and the semantic feature vector of the detected luminous steering wheel sign are passed through a feature principal component significance fusion network to obtain a steering wheel luminous sign multi-dimensional detection feature significance fusion representation vector; Step S1046, respectively passing the reference steering wheel luminous sign image through the HOG feature extractor and the steering wheel luminous sign semantic feature extractor based on the hole convolutional neural network, and inputting the obtained reference steering wheel luminous sign HOG feature vector and reference steering wheel luminous sign semantic feature vector into the feature principal component significance fusion network to obtain a steering wheel luminous sign multi-dimensional reference feature significance fusion representation vector; Step S1047, the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector are passed through a local fine-grained sequence energy dense interactive matching network based on feature decoupling to obtain a steering wheel luminous sign detection-reference semantic matching coefficient; Step S1048: Obtain the quality detection result based on the detection of the steering wheel illuminated sign and the reference semantic matching coefficient.

[0039] Accordingly, in the quality inspection of the front steering wheel luminous sign to obtain the quality inspection result, the detection steering wheel luminous sign image of the front steering wheel luminous sign is collected by the camera, and the reference steering wheel luminous sign image that has passed the quality inspection is obtained from the database, and the image recognition and analysis technology based on deep learning is used to extract the HOG features and semantic features of the detection and reference steering wheel luminous sign images, and the main component is significantly fused, so as to automatically generate the quality inspection result according to the comparison between the semantic matching coefficient between the detection and reference multi-dimensional fusion features and the predetermined threshold. In this way, the influence of light can be eliminated to effectively identify the defects on the sign, while reducing the interference of human factors, ensuring the consistency and reliability of the inspection results, and realizing the intelligent quality inspection of the front steering wheel luminous sign.

[0040] Specifically, in the technical solution of the present application, first, in the quality inspection of the front steering wheel luminous sign to obtain the quality inspection result, the detection steering wheel luminous sign image of the front steering wheel luminous sign is collected by a camera, and the reference steering wheel luminous sign image that has passed the quality inspection is obtained from the database, and the image recognition and analysis technology based on deep learning is used to extract the HOG features and semantic features of the detection and reference steering wheel luminous sign images, and the main component of the two features is significantly fused respectively, so as to automatically obtain the quality inspection result according to the comparison between the semantic matching coefficient between the detection and reference multi-dimensional fusion features and the predetermined threshold. Next, considering that the detection steering wheel luminous sign image contains the edge and texture feature information of the sign, based on this, in the technical solution of the present application, the detection steering wheel luminous sign image is passed through the HOG feature extractor to capture the local shape and texture information of the image to obtain the detection steering wheel luminous sign HOG feature vector.

[0041] Then, considering that the detected steering wheel illuminated sign image also contains contextual semantic information about the sign, based on this, in order to capture more extensive contextual information so as to understand the overall pattern and detail information of the sign, in the technical solution of the present application, the detected steering wheel illuminated sign image is passed through a steering wheel illuminated sign semantic feature extractor based on a hole convolutional neural network to obtain a detected steering wheel illuminated sign semantic feature vector.

[0042] Furthermore, in order to capture and fuse the key and significant features in the HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign, so as to more comprehensively characterize the features of the detected image, in the technical solution of the present application, the HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign are fused through a feature principal component significance network to obtain a steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector. Specifically, the HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign are first standardized to obtain standardized features. After that, the sample covariance matrix of the standardized features is calculated, and the sample covariance matrix is ​​subjected to matrix decomposition-based feature vector extraction to obtain a set of HOG principal component feature vectors of the detected steering wheel luminous sign and a set of semantic principal component feature vectors of the detected steering wheel luminous sign. That is, principal component analysis eliminates the redundant parts of the data through dimensionality reduction technology, while retaining the most representative features, thereby accurately capturing the core elements of the feature vector. Furthermore, based on the maximum approximate query matching between feature vectors, the best matching pair is found by calculating and comparing the similarities between sets of different feature vectors. Then, the best matching pair of each feature vector is subjected to local semantic fine-grained association analysis based on the gating mechanism, so as to enrich the significant semantic association between each set of matching pairs at the local fine-grained level and obtain a set of principal component fusion feature vectors. Finally, the set of principal component fusion feature vectors is cascaded to obtain the final steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector.

[0043] In one embodiment of the present application, the HOG feature vector of the steering wheel luminous sign detection and the semantic feature vector of the steering wheel luminous sign detection are passed through a feature principal component significance fusion network to obtain a steering wheel luminous sign multi-dimensional detection feature significance fusion representation vector, including: standardizing the steering wheel luminous sign detection HOG feature vector and the steering wheel luminous sign detection semantic feature vector to obtain a standardized steering wheel luminous sign detection HOG feature vector and a standardized steering wheel luminous sign detection semantic feature vector; calculating the sample covariance matrix of the standardized steering wheel luminous sign detection HOG feature vector and the standardized steering wheel luminous sign detection semantic feature vector to obtain a steering wheel luminous sign detection HOG sample covariance matrix and a steering wheel luminous sign detection semantic sample covariance matrix; performing matrix decomposition-based feature vector extraction on the steering wheel luminous sign detection HOG sample covariance matrix and the steering wheel luminous sign detection semantic sample covariance matrix to obtain a steering wheel luminous sign detection HOG principal component A set of HOG principal component feature vectors for detecting steering wheel luminous signs and a set of semantic principal component feature vectors for detecting steering wheel luminous signs; input the set of HOG principal component feature vectors for detecting steering wheel luminous signs and the set of semantic principal component feature vectors for detecting steering wheel luminous signs into a maximum approximate query matching network to obtain a set of best matching pairs of HOG principal component feature vectors for detecting steering wheel luminous signs and semantic principal component feature vectors for detecting steering wheel luminous signs; input each best matching pair of HOG principal component feature vectors for detecting steering wheel luminous signs and semantic principal component feature vectors for detecting steering wheel luminous signs in the set of best matching pairs of HOG principal component feature vectors for detecting steering wheel luminous signs and semantic principal component feature vectors for detecting steering wheel luminous signs into a semantic fine-grained gated joint module to obtain a set of HOG-semantic principal component fusion feature vectors for detecting steering wheel luminous signs; cascade the set of HOG-semantic principal component fusion feature vectors for detecting steering wheel luminous signs to obtain a significant fusion representation vector of multi-dimensional detection features of the steering wheel luminous sign.

[0044] Further, in one embodiment of the present application, the HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign are standardized to obtain a standardized HOG feature vector of the detected steering wheel luminous sign and a standardized semantic feature vector of the detected steering wheel luminous sign, including: respectively calculating the mean and standard deviation of the HOG feature vector of the detected steering wheel luminous sign to obtain the HOG feature mean of the detected steering wheel luminous sign and the HOG feature standard deviation of the detected steering wheel luminous sign; subtracting the HOG feature vector of the detected steering wheel luminous sign from the HOG feature mean of the detected steering wheel luminous sign by position, and calculating the HOG feature vector of the detected steering wheel luminous sign The OG difference vector and the standard deviation of the HOG feature of the detected steering wheel luminous sign are divided by position to obtain the standardized HOG feature vector of the detected steering wheel luminous sign; the mean and standard deviation of the semantic feature vector of the detected steering wheel luminous sign are calculated respectively to obtain the mean of the semantic features of the detected steering wheel luminous sign and the standard deviation of the semantic features of the detected steering wheel luminous sign; after subtracting the semantic feature vector of the detected steering wheel luminous sign from the mean of the semantic features of the detected steering wheel luminous sign by position, the calculated semantic offset vector of the detected steering wheel luminous sign is divided by the standard deviation of the semantic features of the detected steering wheel luminous sign to obtain the standardized semantic feature vector of the detected steering wheel luminous sign.

[0045] Further, in one embodiment of the present application, the sample covariance matrix of the standardized detection steering wheel luminous sign HOG feature vector and the standardized detection steering wheel luminous sign semantic feature vector is calculated to obtain the detection steering wheel luminous sign HOG sample covariance matrix and the detection steering wheel luminous sign semantic sample covariance matrix, including: multiplying the transposed vector of the standardized detection steering wheel luminous sign HOG feature vector with the standardized detection steering wheel luminous sign HOG feature vector, and then dividing the obtained standardized detection steering wheel luminous sign HOG association matrix by the value obtained by subtracting one from the length of the standardized detection steering wheel luminous sign HOG feature vector by position to obtain the detection steering wheel luminous sign HOG sample covariance matrix; multiplying the transposed vector of the standardized detection steering wheel luminous sign semantic feature vector with the standardized detection steering wheel luminous sign semantic feature vector, and then dividing the obtained standardized detection steering wheel luminous sign semantic association matrix by the value obtained by subtracting one from the length of the standardized detection steering wheel luminous sign semantic feature vector by position to obtain the detection steering wheel luminous sign semantic sample covariance matrix.

[0046] Furthermore, in one embodiment of the present application, the set of the HOG principal component feature vectors for detecting the luminous sign on the steering wheel and the set of the semantic principal component feature vectors for detecting the luminous sign on the steering wheel are input into a maximum approximate query matching network to obtain a set of the best matching pairs of the HOG principal component feature vectors for detecting the luminous sign on the steering wheel and the semantic principal component feature vectors for detecting the luminous sign on the steering wheel, including: extracting a predetermined HOG principal component feature vector for detecting the luminous sign on the steering wheel from the set of the HOG principal component feature vectors for detecting the luminous sign on the steering wheel; calculating the predetermined HOG principal component feature vector for detecting the luminous sign on the steering wheel; The cosine similarity between the principal component feature vector and each semantic principal component feature vector of the detected steering wheel luminous sign in the set of semantic principal component feature vectors of the detected steering wheel luminous sign is used to obtain a set of matching query similarities; the semantic principal component feature vector of the detected steering wheel luminous sign corresponding to the largest matching query similarity in the set of matching query similarities and the predetermined detected steering wheel luminous sign HOG principal component feature vector are taken as the best matching pair of the predetermined detected steering wheel luminous sign HOG principal component feature vector and the semantic principal component feature vector of the detected steering wheel luminous sign.

[0047] Furthermore, in one embodiment of the present application, each best matching pair of the HOG principal component feature vector for detecting the steering wheel luminous sign and the semantic principal component feature vector for detecting the steering wheel luminous sign is input into the semantic fine-grained gated joint module to obtain a set of HOG-semantic principal component fusion feature vectors for detecting the steering wheel luminous sign, including: respectively calculating the position difference, position dot multiplication and position addition between the best matching pairs of the HOG principal component feature vector for detecting the steering wheel luminous sign and the semantic principal component feature vector for detecting the steering wheel luminous sign to obtain the detection direction The HOG-semantic difference vector of the detected steering wheel luminous sign, the HOG-semantic dot product vector of the detected steering wheel luminous sign and the HOG-semantic sum vector of the detected steering wheel luminous sign; the HOG-semantic difference vector of the detected steering wheel luminous sign, the HOG-semantic dot product vector of the detected steering wheel luminous sign and the HOG-semantic sum vector of the detected steering wheel luminous sign are cascaded and then one-dimensional convolutional encoding is performed to obtain a multi-dimensional fusion vector of the HOG-semantic principal component pattern of the detected steering wheel luminous sign; the multi-dimensional fusion vector of the HOG-semantic principal component pattern of the detected steering wheel luminous sign is subjected to maximum pooling processing based on a local window to obtain the HOG-semantic principal component fusion feature vector of the detected steering wheel luminous sign.

[0048] Specifically, the HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign are processed through a feature principal component significance fusion network using the following significant fusion formula to obtain a significant fusion representation vector of the multi-dimensional detection feature of the steering wheel luminous sign; wherein the significant fusion formula is:

[0049] in, and Respectively represent the HOG feature vector of the detected luminous sign on the steering wheel and the semantic feature vector of the detected luminous sign on the steering wheel, and are respectively the means of the HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign, and are the standard deviations of the HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign, respectively. and are respectively the HOG feature vector of the standardized detection of the luminous sign on the steering wheel and the semantic feature vector of the standardized detection of the luminous sign on the steering wheel, and They are and The transposed vector of and are respectively the number of eigenvalues ​​in the standardized detection steering wheel luminous sign HOG feature vector and the standardized detection steering wheel luminous sign semantic feature vector, and are respectively the HOG sample covariance matrix of the detected steering wheel luminous sign and the semantic sample covariance matrix of the detected steering wheel luminous sign, and They are the HOG principal component orthogonal matrix for detecting the luminous sign on the steering wheel and the orthogonal matrix for detecting the semantic principal component of the luminous sign on the steering wheel. and They are the HOG diagonal matrix for detecting the luminous sign on the steering wheel and the semantic diagonal matrix for detecting the luminous sign on the steering wheel. The diagonal elements of the matrix are The HOG diagonal matrix of the detected steering wheel luminous sign, are the weight values ​​of the HOG principal component feature vectors of each detected steering wheel luminous sign, The diagonal elements of the matrix are The semantic diagonal matrix of the detected steering wheel light sign, are the weight values ​​of the semantic principal component feature vectors of each detected steering wheel luminous sign, and They are and The transposed matrix of is each HOG principal component feature vector of the luminous steering wheel sign in the set of HOG principal component feature vectors of the luminous steering wheel sign, is each semantic principal component feature vector of the steering wheel luminous sign detection in the set of semantic principal component feature vectors of the steering wheel luminous sign detection, To calculate the and The vector inner product between To calculate the one-norm of a vector, To return the maximum value value, is the maximum approximate matching value, , and They are position difference, position dot multiplication and position addition respectively. For cascade processing, is a one-dimensional convolutional coding operation, is the maximum pooling operation, It is the first in the set of HOG-semantic principal component fusion feature vectors for detecting the luminous sign of the steering wheel. HOG-semantic principal component fusion feature vectors for detecting the illuminated sign on the steering wheel, is the number of feature vectors in the set of HOG-semantic principal component fusion feature vectors for detecting the luminous sign on the steering wheel, It is the significant fusion representation vector of the multi-dimensional detection features of the steering wheel luminous sign.

[0050] Similarly, considering that the reference steering wheel luminous sign image also contains edge and texture features, and there is also a semantic association between contexts, the reference steering wheel luminous sign image is passed through the HOG feature extractor and the steering wheel luminous sign semantic feature extractor based on the hole convolutional neural network. Then, in order to further capture and refine the significant features in the entire image, the obtained reference steering wheel luminous sign HOG feature vector and the reference steering wheel luminous sign semantic feature vector are input into the feature principal component significance fusion network to obtain the steering wheel luminous sign multi-dimensional reference feature significance fusion representation vector. The processing method of the reference steering wheel luminous sign image is consistent with the processing method of the detection steering wheel luminous sign image.

[0051] Furthermore, in order to more accurately and effectively, fine-grainedly process and compare the semantic differences between reference and detection features, so as to provide powerful decision support for quality inspection. In the technical solution of the present application, the multi-dimensional detection feature significant fusion representation vector of the steering wheel luminous sign and the multi-dimensional reference feature significant fusion representation vector of the steering wheel luminous sign are significantly fused through a local fine-grained sequence energy-intensive interactive matching network based on feature decoupling to obtain the steering wheel luminous sign detection-reference semantic matching coefficient. Specifically, the detection and reference feature significant fusion representation vectors are first feature decoupled to provide a basis for subsequent more fine-grained local analysis. Then, feature selection is performed on the features obtained after decoupling to eliminate noise and non-critical features and retain only the most discernible features. Then, the energy synergy response value between the selected feature vector pairs is calculated to quantify the local synergy effect and fine-grained semantic association between feature vectors, forming a feature synergy energy fine-grained global distribution matrix. Subsequently, the global distribution matrix is ​​subjected to hole convolution coding to capture and extract deeper associated features to obtain a fine-grained global distribution feature matrix. Finally, the fine-grained global distribution feature matrix is ​​input into the inter-sequence matching metric network to calculate the matching similarity, so as to provide a semantic similarity metric between the reference and detection image features, and obtain the steering wheel light sign detection-reference semantic matching coefficient. In this way, the feature details of the local area between the two can be captured, which is very helpful for understanding the complex patterns and textures of the steering wheel light sign. And through dense interactive matching, the network can accurately align and compare features at the local level, thereby obtaining a more accurate matching coefficient.

[0052] In one embodiment of the present application, the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector are passed through a local fine-grained sequence energy-intensive interactive matching network based on feature decoupling to obtain a steering wheel luminous sign detection-reference semantic matching coefficient, including: feature decoupling the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector to obtain a sequence of steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and a sequence of steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors; feature selection is performed on the sequence of steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and the sequence of steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors to obtain a selected sequence of steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and a selected sequence of steering wheel luminous sign multi-dimensional A sequence of reference feature local significant fusion representation vectors; calculating the energy synergy response value between any pair of selected steering wheel luminous sign multidimensional detection feature local significant fusion representation vectors and the selected steering wheel luminous sign multidimensional reference feature local significant fusion representation vectors in the sequence of selected steering wheel luminous sign multidimensional detection feature local significant fusion representation vectors and the sequence of selected steering wheel luminous sign multidimensional reference feature local significant fusion representation vectors to obtain a steering wheel luminous sign multidimensional reference-detection feature synergy energy fine-grained global distribution matrix; performing atrous convolution encoding on the steering wheel luminous sign multidimensional reference-detection feature synergy energy fine-grained global distribution matrix to obtain a steering wheel luminous sign multidimensional reference-detection feature synergy energy fine-grained global distribution feature matrix; inputting the steering wheel luminous sign multidimensional reference-detection feature synergy energy fine-grained global distribution feature matrix into a sequence matching degree measurement network to obtain the steering wheel luminous sign detection-reference semantic matching coefficient.

[0053] Furthermore, in one embodiment of the present application, feature selection is performed on the sequence of the steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and the sequence of the steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors to obtain the selected sequence of the steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and the selected sequence of the steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors, including: using a first importance measurement function to calculate the steering wheel luminous sign multi-dimensional detection importance factor of each steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vector in the sequence of the steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vector; based on the relationship between each of the steering wheel luminous sign multi-dimensional detection importance factors and a first preset threshold , perform feature selection on the sequence of local significant fusion representation vectors of the multidimensional detection features of the steering wheel luminous sign to obtain the selected sequence of local significant fusion representation vectors of the multidimensional detection features of the steering wheel luminous sign; use the second importance measurement function to calculate the steering wheel luminous sign multidimensional reference importance factor of each steering wheel luminous sign multidimensional reference feature local significant fusion representation vector in the sequence of multidimensional reference feature local significant fusion representation vectors of the steering wheel luminous sign; based on the comparison between each of the multidimensional reference importance factors of the steering wheel luminous sign and the second preset threshold, perform feature selection on the sequence of local significant fusion representation vectors of the multidimensional reference feature of the steering wheel luminous sign to obtain the selected sequence of local significant fusion representation vectors of the multidimensional reference feature of the steering wheel luminous sign.

[0054] Specifically, the first importance metric function is used to calculate the steering wheel luminous sign multidimensional detection importance factor of each steering wheel luminous sign multidimensional detection feature local significant fusion representation vector in the sequence of the steering wheel luminous sign multidimensional detection feature local significant fusion representation vector by the following formula; wherein the calculation formula is:

[0055] in, are respectively the local significant fusion representation vectors of the multi-dimensional detection features of the steering wheel luminous sign in the sequence of the local significant fusion representation vectors of the multi-dimensional detection features of the steering wheel luminous sign, is the detection weight matrix, To detect the bias vector, express function, is the weight coefficient vector, Multi-dimensionally detect importance factors for each steering wheel luminous sign.

[0056] Furthermore, in one embodiment of the present application, the energy synergy response value between any pair of selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors in the sequence of the selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and the selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors is calculated to obtain the steering wheel luminous sign multi-dimensional reference-detection feature synergy energy fine-grained global distribution matrix, including: calculating the selected steering wheel luminous sign multi-dimensional detection feature local significant fusion table The selected steering wheel luminous sign multi-dimensional local area matching feature vector of the selected steering wheel luminous sign is obtained by position-by-position division between the indicator vector and the local significant fusion representation vector of the selected steering wheel luminous sign multi-dimensional reference feature; the maximum value of the selected steering wheel luminous sign multi-dimensional local area matching feature vector is extracted to obtain the maximum value of the selected steering wheel luminous sign multi-dimensional local area matching feature; the mean and variance of the selected steering wheel luminous sign multi-dimensional local area matching feature vector are calculated respectively to obtain the mean of the selected steering wheel luminous sign multi-dimensional local area matching feature and the variance of the selected steering wheel luminous sign multi-dimensional local area matching feature; the selected The value obtained by adding the variance of the multi-dimensional local area matching feature of the selected steering wheel luminous sign and the bias term is multiplied by four to obtain the first energy collaborative response value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign; the square of the difference between the maximum value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign and the mean value of the multi-dimensional local area matching feature of the selected steering wheel luminous sign is calculated to obtain the multi-dimensional local area feature difference value of the selected steering wheel luminous sign; the modulated value obtained by multiplying the variance of the multi-dimensional local area matching feature of the selected steering wheel luminous sign by two is modulated with the variance of the multi-dimensional local area matching feature of the selected steering wheel luminous sign. The value obtained by multiplying the bias term by two and the difference value of the selected multi-dimensional local area feature of the steering wheel luminous sign are added to obtain the second energy synergistic response value of the selected multi-dimensional local area matching feature of the steering wheel luminous sign; the first energy synergistic response value of the selected multi-dimensional local area matching feature of the steering wheel luminous sign is divided by the second energy synergistic response value of the selected multi-dimensional local area matching feature of the steering wheel luminous sign to obtain the energy synergistic response value between the selected local significant fusion representation vector of the multi-dimensional detection feature of the steering wheel luminous sign and the selected local significant fusion representation vector of the multi-dimensional reference feature of the steering wheel luminous sign.

[0057] Specifically, the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector are processed by a local fine-grained sequence energy dense interactive matching network based on feature decoupling using the following inter-sequence matching formula to obtain the steering wheel luminous sign detection-reference semantic matching coefficient; wherein the inter-sequence matching formula is:

[0058] in, and are respectively the multi-dimensional detection feature significant fusion representation vector of the steering wheel luminous sign and the multi-dimensional reference feature significant fusion representation vector of the steering wheel luminous sign, For feature decoupling operations, represents the number of feature vectors in the sequence of the local significant fusion representation vector of the multi-dimensional detection feature of the steering wheel luminous sign and the sequence of the local significant fusion representation vector of the multi-dimensional reference feature of the steering wheel luminous sign, and are respectively a sequence of local significant fusion representation vectors of the multi-dimensional detection features of the steering wheel luminous sign and a sequence of local significant fusion representation vectors of the multi-dimensional reference features of the steering wheel luminous sign, To perform feature selection, and are respectively a sequence of the selected local significant fusion representation vectors of the multi-dimensional detection features of the luminous steering wheel sign and a sequence of the selected local significant fusion representation vectors of the multi-dimensional reference features of the luminous steering wheel sign, and The first and second sequences of the selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vector and the selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vector are respectively represented by The local significant fusion representation vector of the multi-dimensional detection feature of the steering wheel luminous sign after the selection and the The local significant fusion representation vector of the multi-dimensional reference features of the selected steering wheel luminous sign, It means to calculate the energy synergy response value between two vectors. is the first in the fine-grained global distribution matrix of the multi-dimensional reference-detection feature collaborative energy of the steering wheel luminous sign The eigenvalues ​​at the positions, To extract the maximum value of a vector, For the The local significant fusion representation vector of the multi-dimensional detection feature of the steering wheel luminous sign after the selection and the The selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vector between the selected steering wheel luminous sign multi-dimensional local area matching feature vector, and They are The mean and variance of is the bias term, is the multi-dimensional reference-detection feature collaborative energy fine-grained global distribution matrix of the steering wheel luminous sign, is the dilated convolutional coding, is the multi-dimensional reference-detection feature collaborative energy fine-grained global distribution feature matrix of the steering wheel luminous sign, is the matching degree measurement operation between sequences, is the steering wheel illuminated sign detection-reference semantic matching coefficient.

[0059] Then, the steering wheel luminous sign detection-reference semantic matching coefficient is compared with a predetermined threshold to obtain the quality inspection result. That is, the quality inspection result is automatically obtained by comparing the steering wheel luminous sign detection-reference semantic matching coefficient obtained by dense interactive matching of the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector with a predetermined threshold, so as to screen out qualified front steering wheel luminous signs. In this way, the influence of light can be eliminated to effectively identify defects on the sign, while reducing interference from human factors, ensuring the consistency and reliability of the inspection results, and realizing the intelligent quality inspection of the front steering wheel luminous sign.

[0060] In a preferred example, the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector are passed through a local fine-grained sequence energy-intensive interactive matching network based on feature decoupling to obtain the steering wheel luminous sign detection-reference semantic matching coefficient, which includes: Cascading the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector into a steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade representation vector; Determine the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade maximum eigenvalue and the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade minimum eigenvalue of the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade representation vector, and calculate the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade mean and steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade standard deviation of the feature set of the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade representation vector; Calculate the quotient of the steering wheel luminous sign multidimensional detection-reference feature significant fusion cascade mean and the steering wheel luminous sign multidimensional detection-reference feature significant fusion cascade standard deviation to obtain the steering wheel luminous sign multidimensional detection-reference feature significant fusion cascade statistical normalization value; Calculate the inverse of each eigenvalue of the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade representation vector, multiply it by the difference between the maximum eigenvalue of the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade and the minimum eigenvalue of the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade, and then subtract it from the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade statistical normalization value to obtain the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade distribution approximation vector; Calculate an exponential function with a natural constant as the base, with each eigenvalue of the steering wheel luminous sign multidimensional detection-reference feature significant fusion cascade distribution approximation vector as an exponent, and perform point addition with the steering wheel luminous sign multidimensional detection-reference feature significant fusion cascade statistical normalization value to obtain the steering wheel luminous sign multidimensional detection-reference feature significant fusion cascade probability approximation vector; Calculate the base-two logarithm of the absolute value of each eigenvalue of the multi-dimensional detection of the steering wheel light-emitting sign-reference feature significant fusion cascade probability approximation vector to obtain an optimized multi-dimensional detection of the steering wheel light-emitting sign-reference feature significant fusion cascade representation vector; Splitting the optimized steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade representation vector into an optimized steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and an optimized steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector; and The optimized steering wheel illuminated sign multi-dimensional detection feature significant fusion representation vector and the optimized steering wheel illuminated sign multi-dimensional reference feature significant fusion representation vector are passed through a local fine-grained sequence energy-intensive interactive matching network based on feature decoupling to obtain the steering wheel illuminated sign detection-reference semantic matching coefficient.

[0061] Among them, the multi-dimensional detection of the steering wheel luminous sign-reference feature significant fusion cascade representation vector, for example, is recorded as The optimization expression is:

[0062] , , and They are the feature sets of the multi-dimensional detection of the steering wheel luminous sign-reference feature significant fusion cascade representation vector The mean, standard deviation, minimum and maximum values ​​of is the logarithm to base 2, To calculate the inverse of each eigenvalue of the steering wheel luminous sign multi-dimensional detection-reference feature significant fusion cascade representation vector, is the optimized steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector, It is an exponential function that calculates the exponential of each eigenvalue of a vector with a natural constant as the base. is the base-two logarithm, It is added by location point. It is the point multiplication by position. It is subtracted by position point.

[0063] Here, in the preferred example, considering that the steering wheel illuminated sign multi-dimensional detection feature significant fusion representation vector and the steering wheel illuminated sign multi-dimensional reference feature significant fusion representation vector respectively represent the HOG-image semantic fusion features of the detected steering wheel illuminated sign image and the reference steering wheel illuminated sign image, when performing local fine-grained sequence energy-intensive interactive matching based on feature decoupling to map them toward a common semantic matching domain, the difference in image semantic feature fusion distribution under different source image semantics will cause the steering wheel illuminated sign multi-dimensional detection feature significant fusion representation vector and the steering wheel illuminated sign multi-dimensional reference feature significant fusion representation vector to have image semantic sequence interactive matching local fine-grained energy-intensive differences. Therefore, it is expected to improve the balance between the mapping accuracy and completeness of the steering wheel illuminated sign multi-dimensional detection feature significant fusion representation vector and the steering wheel illuminated sign multi-dimensional reference feature significant fusion representation vector when mapping them toward a common semantic matching domain, thereby improving the accuracy of the obtained steering wheel illuminated sign detection-reference semantic matching coefficient.

[0064] Therefore, by performing random statistical normalization on the energy differentiation feature set after the cascade of the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector, the confidence space constructed based on the overall eigenvalues ​​of the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector is approximated by the standardized continuous probability density distribution of the response hypothesis test relative to the individual eigenvalues ​​of the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector, thereby establishing the target reachability of the energy differentiation feature distribution of the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector to a unified mapping target, so as to achieve the balanced executableness between mapping accuracy and mapping completeness in the mapping process of the energy differentiation feature distribution based on the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector, thereby improving the accuracy of the steering wheel luminous sign detection-reference semantic matching coefficient.

[0065] In summary, the above scheme is adopted to collect the detection steering wheel luminous sign image of the front steering wheel luminous sign through the camera, and obtain the reference steering wheel luminous sign image that has passed the quality inspection from the database, and use the image recognition and analysis technology based on deep learning to extract the HOG features and semantic features of the detection and reference steering wheel luminous sign images, and perform the principal component significant fusion, so as to automatically generate the quality inspection results according to the comparison between the semantic matching coefficient between the detection and reference multi-dimensional fusion features and the predetermined threshold. In this way, the influence of light can be eliminated to effectively identify the defects on the sign, while reducing the interference of human factors, ensuring the consistency and reliability of the inspection results, and realizing the intelligent quality inspection of the front steering wheel luminous sign.

[0066] Figure 3 FIG. 1 is a block diagram of a system for implementing a front steering wheel luminous logo according to an exemplary embodiment. Figure 3 As shown, the system 200 includes: A diffusion film pasting module 201 is used to paste the diffusion film on the surface of the light guide plate to obtain a uniform light guide structure; A first clamping module 202, used for clamping the light-guiding uniform structure, the PCBA board and the base to obtain a light-emitting component; The second clamping module 203 is used to clamp the light-emitting assembly with the light-emitting panel to obtain a front steering wheel light-emitting sign; A quality inspection module 204, used to perform a quality inspection on the front steering wheel luminous sign to obtain a quality inspection result, wherein the quality inspection result is used to indicate whether the quality of the front steering wheel luminous sign meets a predetermined requirement; Wherein, the quality detection module 204 includes: A detection steering wheel luminous sign image acquisition unit 2041 is used to acquire a detection steering wheel luminous sign image of the front steering wheel luminous sign through a camera; A reference steering wheel luminous sign image acquisition unit 2042 is used to acquire a reference steering wheel luminous sign image that has passed quality inspection from a database; The HOG feature extraction unit 2043 is used to pass the detected steering wheel light-emitting sign image through a HOG feature extractor to obtain a detected steering wheel light-emitting sign HOG feature vector; A steering wheel luminous sign semantic feature extraction unit 2044 is used to pass the detected steering wheel luminous sign image through a steering wheel luminous sign semantic feature extractor based on a hole convolutional neural network to obtain a detected steering wheel luminous sign semantic feature vector; The feature principal component significance fusion unit 2045 is used to obtain a steering wheel luminous sign multi-dimensional detection feature significance fusion representation vector by passing the HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign through a feature principal component significance fusion network; The steering wheel luminous sign semantic feature extraction unit 2046 is used to pass the reference steering wheel luminous sign image through the HOG feature extractor and the steering wheel luminous sign semantic feature extractor based on the void convolutional neural network respectively, and input the obtained reference steering wheel luminous sign HOG feature vector and reference steering wheel luminous sign semantic feature vector into the feature principal component significance fusion network to obtain the steering wheel luminous sign multi-dimensional reference feature significance fusion representation vector; The local fine-grained sequence energy-intensive interactive matching unit 2047 is used to obtain the steering wheel luminous sign detection-reference semantic matching coefficient by passing the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector through a local fine-grained sequence energy-intensive interactive matching network based on feature decoupling; The quality detection result generating unit 2048 is used to obtain the quality detection result based on the detection of the steering wheel illuminated sign-reference semantic matching coefficient.

[0067] In one embodiment of the present application, Figure 4As shown, the steering wheel luminous sign includes a luminous panel 1, a diffusion film 2, a light guide plate 3, a PCBA 4, a wiring harness assembly 5, and a base 6. Among them, the luminous panel 1 is made of two-color insert injection molding, the diffusion film 2 is used to make the light source uniform, the light guide plate 3 is used to make the light source uniform, and the PCBA 4 includes: a PCB board and a side-emitting LED.

[0068] like Figure 5 As shown, the steering wheel luminous sign is a two-color injection molded plastic part. One color is a transparent light-transmitting PC material, and the other is a black ABS / PC material. The aluminum sheet is embedded inside. The aluminum sheet is fixed to the steering wheel airbag cover for installation. After the airbag is detonated, the luminous sign will not fall off. If there is no black ABS / PC circle on the outside, it is impossible to directly emit white light, because the light source must be installed inside, so it will not affect the light emission. The luminous sign assembly is fixed to the airbag cover by inserting an aluminum sheet, bending it to the back of the airbag cover and pressing it tightly.

[0069] Further, the manufacturing process of the luminous panel 1 includes: first, one-color injection molding, PC transparent; then two-color injection molding PC+ABS, insert aluminum sheet; then, the back is fully sprayed black to produce a middle island high-gloss black effect; then laser engraving the LOGO luminous area; then PVD (AL) and spraying varnish; finally spraying varnish on the A side. It should be understood that the luminous panel 1, the diffusion film 2, the light guide plate 3, the PCBA 4, the harness assembly 5, and the base 6 are assembled to obtain the steering wheel luminous sign.

[0070] like Figure 6 As shown, the light-emitting panel 1 includes a light-emitting surface cover 11, an anti-scattering film 12, a light guide plate 13, and LED 14. LED 14 emits light from the side (120° light-emitting angle) to the light guide plate 13, and the light guide plate 13 conducts the light to the light-emitting area, thereby making the light-emitting area emit light.

[0071] Figure 7 This is a schematic diagram of the light guide plate structure. Locally arranged grooves allow light to diverge, and the brightness and uniformity of the light-emitting area are adjusted through the dot matrix to ensure customer requirements.

[0072] like Figure 8 and 9 As described above, the diffusion film 2 is made of PET transparent material, and white printing is provided at the LED to prevent the generation of bright spots.

[0073] like Fig.10 and Fig.11 As shown, PCBA board 4 uses a 0.5 thick white board with a reflectivity of 50% to increase brightness.

[0074] like Fig.12As described above, the vehicle power supply is input to the positive and negative electrodes of the light-emitting mark, and the backlight brightness is adjusted by the resistor. The PCB double-sided board has a thickness of 0.5mm. Furthermore, ESD capacitors and diodes are added during the design to prevent external shocks such as reverse pressure, high current, and static electricity.

[0075] Reference below Fig.13 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application 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-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig.13 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0076] like Fig.13 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.

[0077] 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 Fig.13 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.

[0078] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a 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 application are executed.

[0079] It should be noted that the computer-readable medium mentioned above in the present application 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 application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, 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.

[0080] 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.

[0081] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0082] Computer program code for performing the operations of the present application 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 the case of 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).

[0083] 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 application. 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 alternative implementations, 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.

[0084] The modules involved in the embodiments described in this application can be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the test parameter acquisition module can also be described as a "module for acquiring device test parameters corresponding to the target device".

[0085] 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.

[0086] In the context of the present application, 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.

[0087] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution 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 this application (but not limited to) by each other to form a technical solution.

[0088] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed 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 construed as limiting the scope of the application. 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.

[0089] 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 of implementation. 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 realizing a front steering wheel luminous mark, characterized in that: include: Paste the diffusion film on the surface of the light guide plate to obtain a uniform light guide structure; The light-guiding uniform structure, the PCBA board and the base are clamped together to obtain a light-emitting component; The light-emitting assembly is clamped with the light-emitting panel to obtain a front steering wheel light-emitting sign; Performing a quality inspection on the front steering wheel luminous sign to obtain a quality inspection result, wherein the quality inspection result is used to indicate whether the quality of the front steering wheel luminous sign meets a predetermined requirement; Wherein, the front steering wheel luminous sign is subjected to quality inspection to obtain a quality inspection result, wherein the quality inspection result is used to indicate whether the quality of the front steering wheel luminous sign meets the predetermined requirements, including: A detection steering wheel luminous sign image of the front steering wheel luminous sign collected by a camera; Obtain a reference steering wheel illuminated sign image that has passed quality inspection from a database; The detected steering wheel luminous sign image is passed through a HOG feature extractor to obtain a detected steering wheel luminous sign HOG feature vector; The detected steering wheel luminous sign image is passed through a steering wheel luminous sign semantic feature extractor based on a hole convolutional neural network to obtain a detected steering wheel luminous sign semantic feature vector; The HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign are passed through a feature principal component significance fusion network to obtain a steering wheel luminous sign multi-dimensional detection feature significance fusion representation vector; The reference steering wheel luminous sign image is respectively passed through the HOG feature extractor and the steering wheel luminous sign semantic feature extractor based on the hole convolutional neural network, and the obtained reference steering wheel luminous sign HOG feature vector and reference steering wheel luminous sign semantic feature vector are input into the feature principal component significance fusion network to obtain a steering wheel luminous sign multi-dimensional reference feature significance fusion representation vector; The steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector are passed through a local fine-grained sequence energy dense interactive matching network based on feature decoupling to obtain the steering wheel luminous sign detection-reference semantic matching coefficient; Based on detecting the steering wheel illuminated sign and matching the reference semantics coefficient, the quality detection result is obtained.

2. The method for realizing a front steering wheel luminous mark according to claim 1, characterized in that: The HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign are passed through a feature principal component significance fusion network to obtain a steering wheel luminous sign multi-dimensional detection feature significance fusion representation vector, including: The HOG feature vector of the detected steering wheel luminous sign and the semantic feature vector of the detected steering wheel luminous sign are normalized to obtain a standardized HOG feature vector of the detected steering wheel luminous sign and a standardized semantic feature vector of the detected steering wheel luminous sign; Calculate the sample covariance matrix of the standardized HOG feature vector for detecting the luminous steering wheel sign and the standardized semantic feature vector for detecting the luminous steering wheel sign to obtain the HOG sample covariance matrix for detecting the luminous steering wheel sign and the semantic sample covariance matrix for detecting the luminous steering wheel sign; Performing feature vector extraction based on matrix decomposition on the HOG sample covariance matrix of the detection steering wheel luminous sign and the semantic sample covariance matrix of the detection steering wheel luminous sign to obtain a set of HOG principal component feature vectors of the detection steering wheel luminous sign and a set of semantic principal component feature vectors of the detection steering wheel luminous sign; Input the set of HOG principal component feature vectors for detecting the luminous sign on the steering wheel and the set of semantic principal component feature vectors for detecting the luminous sign on the steering wheel into a maximum approximate query matching network to obtain a set of best matching pairs of the HOG principal component feature vectors for detecting the luminous sign on the steering wheel and the semantic principal component feature vectors for detecting the luminous sign on the steering wheel; Input each best matching pair of the HOG principal component feature vector for detecting the luminous steering wheel sign and the semantic principal component feature vector for detecting the luminous steering wheel sign into the semantic fine-grained gated joint module to obtain a set of HOG-semantic principal component fusion feature vectors for detecting the luminous steering wheel sign; The set of HOG-semantic principal component fusion feature vectors for detecting the illuminated steering wheel sign is cascaded to obtain a significant fusion representation vector of the multi-dimensional detection feature of the illuminated steering wheel sign.

3. The method for realizing a front steering wheel luminous mark according to claim 2, characterized in that: The HOG feature vector for detecting the luminous sign on the steering wheel and the semantic feature vector for detecting the luminous sign on the steering wheel are normalized to obtain a standardized HOG feature vector for detecting the luminous sign on the steering wheel and a standardized semantic feature vector for detecting the luminous sign on the steering wheel, including: Respectively calculating the mean and standard deviation of the HOG feature vector for detecting the luminous sign on the steering wheel to obtain the mean of the HOG feature for detecting the luminous sign on the steering wheel and the standard deviation of the HOG feature for detecting the luminous sign on the steering wheel; After subtracting the detected steering wheel luminous sign HOG feature vector from the detected steering wheel luminous sign HOG feature mean by position, the calculated detected steering wheel luminous sign HOG difference vector and the detected steering wheel luminous sign HOG feature standard deviation are divided by position to obtain the standardized detected steering wheel luminous sign HOG feature vector; Respectively calculating the mean and standard deviation of the semantic feature vector of the detection steering wheel luminous sign to obtain the mean of the semantic feature of the detection steering wheel luminous sign and the standard deviation of the semantic feature of the detection steering wheel luminous sign; After subtracting the semantic feature vector of the detected steering wheel luminous sign from the mean of the semantic features of the detected steering wheel luminous sign by position, the calculated semantic offset vector of the detected steering wheel luminous sign and the standard deviation of the semantic features of the detected steering wheel luminous sign are divided by position to obtain the standardized semantic feature vector of the detected steering wheel luminous sign.

4. The method for realizing a front steering wheel luminous mark according to claim 3, characterized in that: Calculating the sample covariance matrix of the standardized HOG feature vector for detecting the luminous steering wheel sign and the standardized semantic feature vector for detecting the luminous steering wheel sign to obtain the HOG sample covariance matrix for detecting the luminous steering wheel sign and the semantic sample covariance matrix for detecting the luminous steering wheel sign, including: After multiplying the transposed vector of the standardized detection steering wheel luminous sign HOG feature vector by the standardized detection steering wheel luminous sign HOG feature vector, the obtained standardized detection steering wheel luminous sign HOG association matrix is ​​divided by the value obtained by subtracting one from the length of the standardized detection steering wheel luminous sign HOG feature vector to obtain the detection steering wheel luminous sign HOG sample covariance matrix; After multiplying the transposed vector of the standardized detection steering wheel luminous sign semantic feature vector by the standardized detection steering wheel luminous sign semantic feature vector, the obtained standardized detection steering wheel luminous sign semantic association matrix is ​​divided by the value obtained by subtracting one from the length of the standardized detection steering wheel luminous sign semantic feature vector to obtain the detection steering wheel luminous sign semantic sample covariance matrix.

5. The method for realizing a front steering wheel luminous mark according to claim 4, characterized in that: Inputting the set of the HOG principal component feature vectors for detecting the luminous sign on the steering wheel and the set of the semantic principal component feature vectors for detecting the luminous sign on the steering wheel into a maximum approximate query matching network to obtain a set of the best matching pairs of the HOG principal component feature vectors for detecting the luminous sign on the steering wheel and the semantic principal component feature vectors for detecting the luminous sign on the steering wheel, including: Extracting a predetermined HOG principal component feature vector for detecting a light-emitting steering wheel sign from the set of HOG principal component feature vectors for detecting a light-emitting steering wheel sign; Calculating the cosine similarity between the predetermined HOG principal component feature vector for detecting the luminous steering wheel sign and each semantic principal component feature vector for detecting the luminous steering wheel sign in the set of semantic principal component feature vectors for detecting the luminous steering wheel sign to obtain a set of matching query similarities; The semantic principal component feature vector of the detected steering wheel luminous sign corresponding to the largest matching query similarity in the set of matching query similarities and the predetermined HOG principal component feature vector of the detected steering wheel luminous sign are taken as the best matching pair of the predetermined HOG principal component feature vector of the detected steering wheel luminous sign and the semantic principal component feature vector of the detected steering wheel luminous sign.

6. The method for realizing a front steering wheel luminous mark according to claim 5, characterized in that: Inputting each best matching pair of the HOG principal component feature vector for detecting the luminous steering wheel sign and the semantic principal component feature vector for detecting the luminous steering wheel sign into the semantic fine-grained gated joint module to obtain a set of HOG-semantic principal component fusion feature vectors for detecting the luminous steering wheel sign, including: Respectively calculating the position difference, position dot product and position addition between the best matching pair of the HOG principal component feature vector for detecting the luminous steering wheel sign and the semantic principal component feature vector for detecting the luminous steering wheel sign to obtain the HOG-semantic difference vector for detecting the luminous steering wheel sign, the HOG-semantic dot product vector for detecting the luminous steering wheel sign and the HOG-semantic sum vector for detecting the luminous steering wheel sign; The HOG-semantic difference vector of the detected steering wheel luminous sign, the HOG-semantic dot product vector of the detected steering wheel luminous sign and the HOG-semantic sum vector of the detected steering wheel luminous sign are cascaded and then one-dimensional convolutional coding is performed to obtain a multi-dimensional fusion vector of the HOG-semantic principal component pattern of the detected steering wheel luminous sign; The multi-dimensional fusion vector of the HOG-semantic principal component pattern of the detected steering wheel luminous sign is subjected to local window-based maximum pooling processing to obtain the HOG-semantic principal component fusion feature vector of the detected steering wheel luminous sign.

7. The method for realizing a front steering wheel luminous mark according to claim 6, characterized in that: The steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector are passed through a local fine-grained sequence energy dense interactive matching network based on feature decoupling to obtain a steering wheel luminous sign detection-reference semantic matching coefficient, including: Performing feature decoupling on the steering wheel luminous sign multi-dimensional detection feature significant fusion representation vector and the steering wheel luminous sign multi-dimensional reference feature significant fusion representation vector to obtain a sequence of steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and a sequence of steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors; Performing feature selection on the sequence of local significant fusion representation vectors of the multi-dimensional detection features of the steering wheel luminous sign and the sequence of local significant fusion representation vectors of the multi-dimensional reference features of the steering wheel luminous sign to obtain a selected sequence of local significant fusion representation vectors of the multi-dimensional detection features of the steering wheel luminous sign and a selected sequence of local significant fusion representation vectors of the multi-dimensional reference features of the steering wheel luminous sign; Calculate the energy synergistic response value between any pair of selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors in the sequence of the selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and the sequence of the selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors to obtain a steering wheel luminous sign multi-dimensional reference-detection feature synergistic energy fine-grained global distribution matrix; Performing hole convolution coding on the steering wheel luminous sign multi-dimensional reference-detection feature collaborative energy fine-grained global distribution matrix to obtain the steering wheel luminous sign multi-dimensional reference-detection feature collaborative energy fine-grained global distribution feature matrix; The multi-dimensional reference-detection feature collaborative energy fine-grained global distribution feature matrix of the steering wheel illuminated sign is input into the sequence matching metric network to obtain the steering wheel illuminated sign detection-reference semantic matching coefficient.

8. The method for realizing a front steering wheel luminous mark according to claim 7, characterized in that: The method comprises: performing feature selection on the sequence of the local significant fusion representation vectors of the multi-dimensional detection features of the steering wheel luminous sign and the sequence of the local significant fusion representation vectors of the multi-dimensional reference features of the steering wheel luminous sign to obtain a sequence of the selected local significant fusion representation vectors of the multi-dimensional detection features of the steering wheel luminous sign and a sequence of the selected local significant fusion representation vectors of the multi-dimensional reference features of the steering wheel luminous sign, including: Calculating the steering wheel luminous sign multidimensional detection importance factor of each steering wheel luminous sign multidimensional detection feature local significant fusion representation vector in the sequence of the steering wheel luminous sign multidimensional detection feature local significant fusion representation vector using a first importance measurement function; Based on the comparison between each of the multi-dimensional detection importance factors of the luminous steering wheel sign and the first preset threshold, feature selection is performed on the sequence of the local significant fusion representation vectors of the multi-dimensional detection features of the luminous steering wheel sign to obtain the selected sequence of the local significant fusion representation vectors of the multi-dimensional detection features of the luminous steering wheel sign; Calculating the steering wheel luminous sign multidimensional reference importance factor of each steering wheel luminous sign multidimensional reference feature local significant fusion representation vector in the sequence of the steering wheel luminous sign multidimensional reference feature local significant fusion representation vector using a second importance measurement function; Based on the comparison between each of the multidimensional reference importance factors of the steering wheel illuminated sign and the second preset threshold, feature selection is performed on the sequence of local significant fusion representation vectors of the multidimensional reference features of the steering wheel illuminated sign to obtain the selected sequence of local significant fusion representation vectors of the multidimensional reference features of the steering wheel illuminated sign.

9. The method for realizing a front steering wheel luminous mark according to claim 8, characterized in that: Calculating the energy synergistic response value between any pair of selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors in the sequence of the selected steering wheel luminous sign multi-dimensional detection feature local significant fusion representation vectors and the sequence of the selected steering wheel luminous sign multi-dimensional reference feature local significant fusion representation vectors to obtain the steering wheel luminous sign multi-dimensional reference-detection feature synergistic energy fine-grained global distribution matrix, including: Calculate the position-by-position division between the selected local significant fusion representation vector of the multi-dimensional detection feature of the luminous steering wheel sign and the selected local significant fusion representation vector of the multi-dimensional reference feature of the luminous steering wheel sign to obtain the selected multi-dimensional local area matching feature vector of the luminous steering wheel sign; Extracting the maximum value of the multi-dimensional local area matching feature vector of the selected luminous steering wheel sign to obtain the maximum value of the multi-dimensional local area matching feature of the selected luminous steering wheel sign; Respectively calculating the mean and variance of the selected multi-dimensional local area matching feature vector of the luminous steering wheel sign to obtain the selected multi-dimensional local area matching feature mean of the luminous steering wheel sign and the selected multi-dimensional local area matching feature variance of the luminous steering wheel sign; The value obtained by adding the selected steering wheel luminous sign multi-dimensional local area matching feature variance and the bias term is multiplied by four to obtain the first energy collaborative response value of the selected steering wheel luminous sign multi-dimensional local area matching feature; Calculate the square of the difference between the maximum value of the selected multi-dimensional local area matching feature of the luminous steering wheel sign and the mean value of the selected multi-dimensional local area matching feature of the luminous steering wheel sign to obtain the multi-dimensional local area feature difference value of the luminous steering wheel sign after selection; The second energy collaborative response value of the selected steering wheel luminous sign multidimensional local area matching feature is obtained by multiplying the selected steering wheel luminous sign multidimensional local area matching feature variance by two, adding the value obtained by multiplying the bias term by two, and the selected steering wheel luminous sign multidimensional local area feature difference value to obtain the selected steering wheel luminous sign multidimensional local area matching feature second energy collaborative response value; Divide the first energy synergistic response value of the selected multi-dimensional local area matching feature of the steering wheel luminous sign by the second energy synergistic response value of the selected multi-dimensional local area matching feature of the steering wheel luminous sign to obtain the energy synergistic response value between the selected local significant fusion representation vector of the multi-dimensional detection feature of the steering wheel luminous sign and the selected local significant fusion representation vector of the multi-dimensional reference feature of the steering wheel luminous sign.

10. The method for realizing a front steering wheel luminous mark according to claim 9, characterized in that: Based on the detection of the steering wheel illuminated sign-reference semantic matching coefficient, the quality detection result is obtained, including: comparing the detection of the steering wheel illuminated sign-reference semantic matching coefficient with a predetermined threshold to obtain the quality detection result.

Citation Information

Patent Citations

  • Traffic sign information extracting method based on integration of laser point cloud and image data

    CN107516077A

  • Rolling bearing unknown fault detection method based on multi-modal feature fusion enhancement

    CN117516937A

  • Miter gate welding quality detection method and system based on machine vision

    CN118037741A

  • Power transmission and transformation equipment inspection system and method based on visible light image recognition technology

    CN118230198A

  • New energy automobile damping rod detection system and method based on machine vision

    CN118396980A

Cited By

  • Preparation process of special sealing primer for construction surface with water

    CN119991599A

  • Preparation process of special sealant for construction surfaces with water

    CN119991599B