Automatic monitoring system and method for plane knife lines of automobile parts

Abstract

The invention belongs to the technical field of detection, and provides an automatic monitoring system and method for plane knife lines of automobile parts. The method comprises the following steps: inputting a regional knife pattern feature set extracted from a high-definition image into a classifier, and integrating a plurality of knife pattern probabilities output by the classifier to obtain a first knife pattern probability corresponding to the regional knife pattern feature set; if the first knife pattern probability is lower than the first probability and higher than the second probability, inputting the regional knife pattern feature set, a plurality of other regional knife pattern feature sets with plane knife patterns and the functional attribute information of the automobile part into a depth analysis model, and if the second knife pattern probability output by the depth analysis model is higher than the first probability, outputting a second knife pattern feature set; if so, determining that the plane knife patterns exist in the automobile part region corresponding to the region knife pattern feature set; and performing label marking on the automobile spare part area which is judged to have the plane knife lines. According to the invention, high-efficiency quality inspection and marking of the plane knife patterns on the automobile parts can be realized, and follow-up treatment of the plane knife patterns is assisted.

Description

Technical Field

[0001] The present invention relates to the field of detection technologies, and more particularly, to an automatic monitoring system and method for planar tool marks on automotive parts. Background Art

[0002] In modern automotive manufacturing, the processing precision and quality of automotive parts play a crucial role in the overall performance and safety of vehicles. With the rapid development of the automotive industry, the requirements for the processing technology of automotive parts are also increasing day by day.

[0003] Currently, during the processing of automotive parts, various factors often cause tool marks on the surface of the parts, as shown in the circled area in Figure 1. On the one hand, tool wear, improper selection of cutting parameters, and accuracy problems of the machine tool, etc., will all cause the tool to be unable to remove materials stably and precisely during cutting, thus leaving tool marks on the surface of the parts. For example, after long-term use, the cutting edge of the tool will gradually wear, resulting in unstable cutting force and thus generating tool marks. On the other hand, the characteristics of different automotive part materials, such as aluminum alloy, cast iron, etc., will also affect the probability of tool mark generation. Some materials have uneven hardness and are more likely to have tool marks during processing.

[0004] For sealed and assembled automotive parts, the problem of tool marks on the surface is particularly serious. Sealed and assembled automotive parts, such as engine blocks, transmission housings, etc., their sealing performance is directly related to the normal operation of the vehicle. Once there are tool marks on the surface of the parts, after assembly, it is difficult for the sealing surface to achieve a tight fit, and gas or liquid will leak along the tiny gaps formed by the tool marks. Such leakage will not only reduce the performance of the vehicle, such as causing a decrease in engine power and leakage of transmission fluid, but may also pose safety hazards and threaten the driving safety of the vehicle.

[0005] How to automatically monitor the planar tool marks on automotive parts is a technical problem that needs to be solved. Summary of the Invention

[0006] In view of the above technical problems, the present invention provides an automatic monitoring method, system, electronic device, computer storage medium, and computer program product for planar tool marks on automotive parts.

[0007] The present invention discloses an automatic monitoring method for planar knife marks of auto parts. The method comprises the following steps: Under the illumination of a uniform light source, high-definition images of the auto parts are taken from multiple angles, and knife mark features are extracted from each of the high-definition images to obtain a plurality of regional knife mark feature sets. Among them, each regional knife mark feature set contains a plurality of knife mark feature groups, and each knife mark feature group corresponds to a high-definition image at a certain angle. Each knife mark feature group in the regional knife mark feature set is respectively input into a classifier, and a plurality of knife mark probabilities output by the classifier are integrated to obtain a first knife mark probability corresponding to the regional knife mark feature set. If the first knife mark probability is higher than a first probability, it is determined that there are planar knife marks in the auto part area corresponding to the regional knife mark feature set. If the first knife mark probability is lower than the first probability and higher than a second probability, the regional knife mark feature set, several other regional knife mark feature sets determined to have planar knife marks, and the functional attribute information of the auto part are input into a depth analysis model. If the second knife mark probability output by the depth analysis model is higher than the first probability, it is determined that there are planar knife marks in the auto part area corresponding to the regional knife mark feature set. Labeling is performed on the auto part area determined to have planar knife marks.

[0008] Optionally, the extracting knife mark features from each of the high-definition images to obtain a plurality of regional knife mark feature sets includes: Image preprocessing: performing grayscale processing on each of the collected high-definition images; using Gaussian filtering or median filtering to remove image noise; enhancing the image contrast through histogram equalization to highlight the texture features. Texture area segmentation: using an edge detection algorithm to extract the texture contour, performing morphological operations to connect broken edges to form a complete texture area, and separating the texture area from the background through a region growing algorithm or a threshold segmentation method. Texture feature extraction: extracting the geometric features of the texture, including direction, spacing, depth, and shape; calculating the texture features of the texture; using Fourier transform or wavelet transform to analyze the frequency features of the texture. Integrating the geometric features, texture features, and frequency features of the texture of a single region extracted from each high-definition image into the knife mark feature group, and integrating each knife mark feature group into the regional knife mark feature set corresponding to the single region.

[0009] Optionally, inputting the regional tool mark feature set, several other regional tool mark feature sets for determining the existence of planar tool marks, and the functional attribute information of the automotive part into the in-depth analysis model includes: determining the location of the regional tool mark feature set on the automotive part, and based on this location, determining the target surface area of the automotive part, and determining each other regional tool mark feature set corresponding to the determination of the existence of planar tool marks in the target surface area as an auxiliary regional tool mark feature set; inputting the regional tool mark feature set, each of the auxiliary regional tool mark feature sets, and the functional attribute information of the automotive part into the in-depth analysis model, and the in-depth analysis model outputs the third tool mark probability; wherein, the functional attribute information includes the usage type and installation position of the automotive part; calculating the second tool mark probability based on the third tool mark probability.

[0010] Optionally, calculating the second tool mark probability based on the third tool mark probability includes: extracting the part feature of the automotive part from any one of the high-definition images, and determining whether the location belongs to the inner side or the outer side of the automotive part based on the part feature; if the location belongs to the inner side of the automotive part, configuring a first coefficient; if the location belongs to the outer side of the automotive part, configuring a second coefficient; wherein, the first coefficient is greater than the second coefficient; using the first coefficient or the second coefficient to optimize the third tool mark probability to obtain the second tool mark probability.

[0011] Optionally, determining the target surface area of the automotive part based on this location includes: determining the probability difference between the first tool mark probability of the regional tool mark feature set and the first probability, and matching the surface radius according to the magnitude of the probability difference; wherein, the surface radius is positively correlated with the probability difference; based on this location, determining the target surface area on the surface of the automotive part according to the surface radius.

[0012] The present invention also discloses an automatic monitoring system for the planar knife marks of automotive parts. The system includes a processing device and a storage device. The computer code stored in the storage device is called and executed by the processing device to implement the following steps: Under the illumination of a uniform light source, high-definition images of the automotive parts from multiple angles are captured, and knife mark features are extracted from each of the high-definition images to obtain a plurality of regional knife mark feature sets; wherein, each regional knife mark feature set contains a plurality of knife mark feature groups, and each knife mark feature group corresponds to a high-definition image at a certain angle; each knife mark feature group in the regional knife mark feature set is respectively input into a classifier, and a plurality of knife mark probabilities output by the classifier are integrated to obtain a first knife mark probability corresponding to the regional knife mark feature set; if the first knife mark probability is higher than a first probability, it is determined that there are planar knife marks in the automotive part area corresponding to the regional knife mark feature set; if the first knife mark probability is lower than the first probability and higher than a second probability, the regional knife mark feature set, several other regional knife mark feature sets determined to have planar knife marks, and the functional attribute information of the automotive part are input into a depth analysis model. If the second knife mark probability output by the depth analysis model is higher than the first probability, it is determined that there are planar knife marks in the automotive part area corresponding to the regional knife mark feature set; label marking is performed on the automotive part area determined to have planar knife marks.

[0013] The present invention also discloses an electronic device, including: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, and the processor executes the computer program to implement the method as described in any one of the preceding items.

[0014] The present invention also discloses a computer storage medium, and the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method as described in any one of the preceding items.

[0015] The present invention also discloses a computer program product, and the computer program product contains computer code, and when the computer code is executed by a processor of an electronic device, the method as described in any one of the preceding items is implemented.

[0016] The beneficial effects of the present invention are at least as follows: 1) Through multi-angle high-definition image acquisition and knife mark feature extraction, the surface information of the parts is comprehensively captured, effectively reducing the omission of knife marks caused by shooting limitations, and greatly improving the comprehensiveness of detection.

[0017] 2) Based on the probability determination mechanism of the classifier and the depth analysis model, compared with the traditional detection method, the detection accuracy is significantly improved, the misjudgment rate is reduced, and the knife marks are accurately identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for use in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a schematic diagram of the plane knife marks existing in the automotive parts disclosed in the embodiments of the present invention.

[0020] Figure 2 It is a schematic flowchart of an automatic monitoring method for the plane knife marks of an automotive part disclosed in the embodiments of the present invention.

[0021] Figure 3 It is a schematic structural diagram of an automatic monitoring system for the plane knife marks of an automotive part disclosed in the embodiments of the present invention. Detailed implementation manners

[0022] The following specific embodiments illustrate the implementation manners of the present application. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of them. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present application.

[0023] In addition, the technical features involved in different implementation manners of the present application described below can be combined with each other as long as they do not conflict with each other.

[0024] As Figure 2 shown, in response to the above technical problems, the embodiments of the present invention disclose an automatic monitoring method for the plane knife marks of automotive parts. The method includes the following steps: S100, under the illumination of a uniform light source, capture high-definition images of the automotive parts from multiple angles, and extract the knife mark features from each of the high-definition images to obtain multiple regional knife mark feature sets; wherein, each regional knife mark feature set contains multiple knife mark feature groups, and each knife mark feature group corresponds to a high-definition image at a certain angle.

[0025] In this step, a uniform light source is arranged in the assembly machine of the automotive parts (such as a bushing assembly machine, a sealing machine) or in a dedicated automotive parts quality inspection machine. Under the illumination of the uniform light source, control the high-definition camera to capture high-definition images of the automotive parts from multiple angles.

[0026] For each high-definition image, use computer vision processing methods to identify each area where planar tool marks may exist, and extract tool mark features, thus obtaining multiple regional tool mark feature sets. Each regional tool mark feature set corresponds to an area where planar tool marks may exist, and contains a set of tool mark features corresponding to that area extracted from each high-definition image.

[0027] It should be noted that due to the complex and diverse characteristics of tool marks, extracting features from multiple angles can more comprehensively reflect the tool mark situation, and images from different angles are also beneficial for highlighting tool marks and improving the recognition accuracy of planar tool marks.

[0028] S200: Input each set of tool mark features in the regional tool mark feature set into the classifier respectively, integrate the multiple tool mark probabilities output by the classifier, and obtain the first tool mark probability corresponding to the regional tool mark feature set; if the first tool mark probability is higher than the first probability, it is determined that there are planar tool marks in the automotive parts area corresponding to the regional tool mark feature set.

[0029] In this step, the present invention constructs a classifier based on machine learning algorithms such as support vector machine (SVM) and random forest. This classifier is small in size and has high processing efficiency. The classifier is used to conduct a preliminary and rapid classification and recognition of planar tool marks for each set of tool mark features in the regional tool mark feature set, obtaining a corresponding number of multiple tool mark probabilities. Integrate all tool mark probabilities, for example, calculate the average value, to obtain the first tool mark probability.

[0030] If the first tool mark probability is higher than the first probability, it indicates that after the classifier classifies the sets of tool mark features from different angles, it is determined that the existence of planar tool marks in this area is highly probable, that is, obvious tool marks. At this time, it is determined that there are planar tool marks in this area.

[0031] S300: If the first tool mark probability is lower than the first probability and higher than the second probability, then input the regional tool mark feature set, several other regional tool mark feature sets determined to have planar tool marks, and the functional attribute information of the automotive parts into the in-depth analysis model. If the second tool mark probability output by the in-depth analysis model is higher than the first probability, it is determined that there are planar tool marks in the automotive parts area corresponding to the regional tool mark feature set.

[0032] In this step, there may be some actively designed patterns on the auto parts, such as functional patterns (such as anti-slip patterns, heat dissipation grooves, flow guide grooves, etc.), decorative patterns (such as wire drawing patterns, matte patterns or other decorative textures), and surface treatment effects (surface treatment processes such as sandblasting, electroplating, oxidation, etc. may form effects similar to knife marks on the surface of the parts). These actively designed patterns have a certain visual similarity to the planar knife marks, and it is difficult to determine whether there are knife marks only relying on the preliminary judgment of the classifier. Correspondingly, when the first knife mark probability obtained based on the classifier is between the first probability and the second probability, it is determined as a suspicious knife mark, and the present invention further uses a depth analysis model to further judge this area.

[0033] At this time, the regional knife mark feature set of the suspicious area, the knife mark feature sets of other areas determined to have planar knife marks, and the functional attribute information of the auto parts are input into the depth analysis model. The depth analysis model conducts in-depth analysis by combining these more comprehensive information and outputs the second knife mark probability. If this probability is higher than the first probability, it is determined that there are indeed planar knife marks in the suspicious area. In this way, the accuracy of knife mark detection is improved. Especially for those situations that are difficult to directly judge, by comprehensively analyzing more information, the probability of missed judgment is reduced.

[0034] S400, label the areas of the auto parts determined to have planar knife marks.

[0035] In this step, labeling the areas with planar knife marks facilitates subsequent targeted processing of these areas, such as manual verification and repair, to ensure the quality and safety of the auto parts and prevent problems such as seal leakage caused by knife marks from affecting the normal operation and safe driving of the vehicle.

[0036] The above technical solutions of the present invention have at least the following beneficial technical effects: 1) By collecting multi-angle high-definition images and extracting knife mark features, the surface information of the parts is comprehensively captured, effectively reducing the omission of knife marks caused by shooting limitations and greatly improving the comprehensiveness of detection; 2) Based on the probability determination mechanism of the classifier and the depth analysis model, compared with the traditional detection method, the detection accuracy is significantly improved, the misjudgment rate is reduced, and the knife marks are accurately identified.

[0037] Optionally, extracting the tool mark features from each of the high-definition images to obtain multiple regional tool mark feature sets, including: Image preprocessing: performing grayscale processing on each of the collected high-definition images; removing image noise using Gaussian filtering or median filtering; enhancing the image contrast through histogram equalization to highlight the texture features; Texture region segmentation: using an edge detection algorithm to extract the texture contour, performing morphological operations such as dilation and erosion to connect broken edges to form a complete texture region, and separating the texture region from the background through region growing algorithm or threshold segmentation method; Texture feature extraction: extracting the geometric features of the texture, including direction, spacing, depth, and shape; calculating the texture features of the texture; analyzing the frequency features of the texture using Fourier transform or wavelet transform; integrating the geometric features, texture features, and frequency features of the texture of a single region extracted from each high-definition image into the tool mark feature group, and integrating each tool mark feature group into the regional tool mark feature set corresponding to this single region.

[0038] In this embodiment, the edge detection algorithm for extracting the texture contour is, for example, the Canny operator, and the morphological operations adopted are, for example, dilation, erosion, etc. The texture features of the texture are, for example, the contrast, energy, entropy, etc. of the gray-level co-occurrence matrix (GLCM).

[0039] Optionally, inputting the regional tool mark feature set, several other regional tool mark feature sets determined to have planar tool marks, and the functional attribute information of the automotive part into the depth analysis model, including: determining the location of the regional tool mark feature set on the automotive part, and determining the target surface area of the automotive part based on this location, and determining each other regional tool mark feature set determined to have planar tool marks in the target surface area as the auxiliary regional tool mark feature set; inputting the regional tool mark feature set, each of the auxiliary regional tool mark feature sets, and the functional attribute information of the automotive part into the depth analysis model, and the depth analysis model outputs the third tool mark probability; wherein, the functional attribute information includes the usage type and installation location of the automotive part; calculating the second tool mark probability based on the third tool mark probability.

[0040] In this embodiment, on the surface of the automotive part, the appearance of tool marks is often not an isolated phenomenon, and the tool marks in a specific area may be related to the surrounding areas. Therefore, the present invention uses a depth analysis model to comprehensively analyze the regional tool mark feature set of the suspicious tool marks that need to be analyzed twice and other regional tool mark feature sets in the surrounding areas that have been determined to have planar tool marks, and supplements with the functional attribute information of the automotive part to comprehensively analyze whether the regional tool mark feature set really belongs to planar tool marks.

[0041] The auxiliary area tool mark feature set can provide more peripheral information to help the depth analysis model more comprehensively understand the situation of this area. For example, if there are suspected tool marks in a certain area and similar texture features also exist in its peripheral area, then the probability that the suspected tool mark is a real tool mark is high. While if there are no similar texture features in the peripheral area, then the probability that the suspected tool mark is a real tool mark is lower. In other words, the auxiliary area tool mark feature set has reference value for judging whether the tool marks in this area are normal design textures or tool marks generated by processing. However, the number of uses of the auxiliary area tool mark feature set should be restricted because excessive use will significantly reduce the analysis efficiency of the depth analysis model. Moreover, if the planar tool marks are caused by excessive wear of the machine tool cutter, the planar tool marks usually exist in multiple areas. However, the appearance of the planar tool marks is more related to the automotive parts themselves. For example, the special complex structure of the automotive parts increases the control difficulty of the processing machine tool, resulting in inevitable slight jitter of the cutter of the processing machine tool, thus causing the existence of planar tool marks. In this case, the planar tool marks are actually regularly distributed at certain specific structures. Therefore, only the tool mark feature set corresponding to the surrounding area that needs to be re-analyzed needs to be used as the auxiliary area tool mark feature set, and there is no need to input the tool mark feature sets of all areas on the automotive parts determined to have planar tool marks into the depth analysis model.

[0042] After inputting the tool mark feature set of this area, each auxiliary area tool mark feature set, and the functional attribute information of the automotive parts into the depth analysis model, the depth analysis model can output the third tool mark probability. This third tool mark probability reflects the possibility that there are tool marks in this area determined based on various input information. Finally, the above-mentioned second tool mark probability is calculated based on the third tool mark probability analyzed by the depth analysis model.

[0043] It should be noted that the functional attribute information of automotive parts includes the use type (for sealing, transmission, or other functions), installation location (inside the engine, on the chassis, or on the body, etc.). The possibility of having textures similar to planar tool marks on the surfaces of parts with different uses and installation locations is different. For example, the possibility of having textures similar to planar tool marks on decorative parts is significantly higher than that on the engine cylinder head. This functional attribute information can be used to assist the analysis of the depth analysis model.

[0044] In addition, the deep analysis model in the present invention is preferably built based on a general-purpose large model such as DeepSeek, Chart GPT, etc. For example, the above-mentioned general-purpose large model is locally deployed and retrained, and then embedded into the automatic monitoring system of the present invention. When in use, the knife mark feature set of the area, the knife mark feature sets of several other areas where the plane knife marks are determined to exist, and the functional attribute information of the auto parts are sent to the locally deployed general-purpose large model in the form of questions, and the answer content includes the above-mentioned third knife mark probability. The deep analysis model can also be a vertical model built and trained based on algorithms such as CNN and Transformer, and the present invention does not specifically limit this.

[0045] Optionally, calculating the second knife mark probability based on the third knife mark probability includes: extracting the accessory feature of the auto part based on any of the high-definition images, and determining whether the location belongs to the inside or outside of the auto part based on the accessory feature; if the location belongs to the inside of the auto part, configuring a first coefficient; if the location belongs to the outside of the auto part, configuring a second coefficient; wherein the first coefficient is greater than the second coefficient; and using the first coefficient or the second coefficient to optimize the third knife mark probability to obtain the second knife mark probability.

[0046] In this embodiment, some of the lines actively designed on the auto parts are generally present on the outside of the auto parts, and such lines are generally not designed inside the auto parts. For example, decorative lines are generally designed on the outside of the decorative plate.

[0047] Based on the above-mentioned actual design rules, the present invention further extracts the accessory features of the auto part based on any one of the high-definition images taken above, including the shape of the accessory, the position of the connector (such as the bolt connection column / hole), etc. Based on these features, it can be determined whether the area where the suspicious knife mark is located belongs to the inside or outside of the auto part. For example, for an engine cylinder head with a complex shape, by analyzing its image, it can be clearly determined that the side with the bolt connection column / hole is the inside, and the other side is the outside, thereby determining whether the area where the suspicious knife mark is located is the inside or the outside.

[0048] Based on the above analysis, since the probability of the inner and outer sides of the auto parts being actively designed to be misidentified as flat knife marks is different, the present invention further configures different optimization coefficients for the inner and outer sides of the auto parts. Specifically, when the area where the suspicious knife marks are located belongs to the inner side of the auto parts, a larger first coefficient, such as 1.2, is configured, that is, the recognition sensitivity of the suspicious knife marks on the inner side is improved; when the area where the suspicious knife marks are located belongs to the outer side of the auto parts, a slightly smaller second coefficient, such as 1.0 or 1.1, is configured, that is, the recognition sensitivity of the suspicious knife marks on the outer side is reduced.

[0049] In addition, since the influence of the inner planar tool marks on the function of automotive parts is significantly higher than that of the outer side, a larger first coefficient is configured for the inner side, so that the probability of the inner suspicious tool marks is reasonably amplified during calculation, which also conforms to the actual situation that the inner side is more sensitive to tool marks and the harm of tool marks is greater; correspondingly, a slightly smaller second coefficient is configured for the outer side to appropriately reduce the weight of the outer tool mark probability, which is in line with the relatively higher tolerance of the outer side to tool marks. It should be noted that the above optimization can be the multiplication operation of the first coefficient or the second coefficient and the third tool mark probability.

[0050] Optionally, determining the target surface area of the automotive part based on this location includes: determining the probability difference between the first tool mark probability and the first probability of the tool mark feature set in this area, and matching the surface radius according to the magnitude of the probability difference; wherein, the surface radius is positively correlated with the probability difference; based on this location, the target surface area is determined on the surface of the automotive part according to the surface radius.

[0051] In this embodiment, as described above, the target surface area should be set to an appropriate size so as to determine an appropriate number of other area tool mark feature sets as auxiliary area tool mark feature sets. Specifically: The present invention first calculates the probability difference between the first tool mark probability and the first probability of the suspicious tool mark area, and this probability difference is used to represent the degree of closeness between the first tool mark probability and the first probability. When the degree of closeness is higher, it indicates that this suspicious tool mark area is more like a planar tool mark. At this time, the surface radius is set smaller, that is, only the other area tool mark feature sets of multiple areas recognized as having planar tool marks in a small range are determined as auxiliary area tool mark feature sets, which can reduce the analysis load of the depth analysis model. And when the degree of closeness is lower, it indicates that this suspicious tool mark area is more like an actively designed pattern. At this time, more auxiliary area tool mark feature sets need to be called so that the depth analysis model can use more reference data for more accurate and detailed analysis. Through the above adjustment of the surface radius, the present invention achieves a balance between analysis efficiency and analysis accuracy.

[0052] It should be noted that the above-mentioned surface radius refers to the actual path along the contour of the outer surface of the automotive part, rather than a straight path in a two-dimensional plane.

[0053] Such as Figure 3As shown in the figure, an embodiment of the present invention also discloses an automatic monitoring system for the planar knife marks of automotive parts. The system includes a processing device and a storage device. The computer code stored in the storage device is called and executed by the processing device to implement the following steps: Under the illumination of a uniform light source, capture high-definition images of the automotive parts from multiple angles, extract the knife mark features from each of the high-definition images, and obtain multiple regional knife mark feature sets; wherein, each regional knife mark feature set contains multiple knife mark feature groups, and each knife mark feature group corresponds to a high-definition image at a certain angle; input each knife mark feature group in the regional knife mark feature set into a classifier respectively, integrate the multiple knife mark probabilities output by the classifier, and obtain the first knife mark probability corresponding to the regional knife mark feature set; if the first knife mark probability is higher than the first probability, it is determined that there are planar knife marks in the automotive part area corresponding to the regional knife mark feature set; if the first knife mark probability is lower than the first probability and higher than the second probability, input the regional knife mark feature set, several other regional knife mark feature sets determined to have planar knife marks, and the functional attribute information of the automotive part into a depth analysis model. If the second knife mark probability output by the depth analysis model is higher than the first probability, it is determined that there are planar knife marks in the automotive part area corresponding to the regional knife mark feature set; perform label marking on the automotive part area determined to have planar knife marks.

[0054] An embodiment of the present invention also discloses an electronic device, including: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor. The processor executes the computer program to implement the method as described in the foregoing embodiment.

[0055] An embodiment of the present invention also discloses a computer storage medium. The computer storage medium stores a computer program, and the computer program is executed by a processor to implement the method as described in the foregoing embodiment.

[0056] An embodiment of the present invention also discloses a computer program product. The computer program product contains computer code, and when the computer code is executed by the processor of an electronic device, it implements the method as described in the foregoing embodiment.

[0057] The computer-readable storage medium described above may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0058] In order to provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, speech input, or tactile input).

[0059] It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention may be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is imposed herein.

[0060] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An automatic monitoring method for the flat knife marks of automotive parts, characterized in that: The method includes the following steps: Under the illumination of a uniform light source, capture high-definition images of the automotive parts from multiple angles, extract the knife mark features from each of the high-definition images to obtain multiple regional knife mark feature sets; wherein, each regional knife mark feature set contains multiple knife mark feature groups, and each knife mark feature group corresponds to a high-definition image at a certain angle; input each knife mark feature group in the regional knife mark feature set into a classifier respectively, integrate the multiple knife mark probabilities output by the classifier to obtain the first knife mark probability corresponding to this regional knife mark feature set; if the first knife mark probability is higher than the first probability, it is determined that there are planar knife marks in the automotive part area corresponding to this regional knife mark feature set; if the first knife mark probability is lower than the first probability and higher than the second probability, input this regional knife mark feature set, several other regional knife mark feature sets determined to have planar knife marks, and the functional attribute information of this automotive part into a depth analysis model, if the second knife mark probability output by the depth analysis model is higher than the first probability, it is determined that there are planar knife marks in the automotive part area corresponding to this regional knife mark feature set; perform label marking on the automotive part areas determined to have planar knife marks.

2. The automatic monitoring method for the flat knife marks of an automotive accessory according to claim 1, characterized in that: The extracting the knife mark features from each of the high-definition images to obtain multiple regional knife mark feature sets includes: Image preprocessing: perform grayscale processing on each of the collected high-definition images; use Gaussian filtering or median filtering to remove image noise; enhance the image contrast through histogram equalization to highlight the texture features; Texture area segmentation: use an edge detection algorithm to extract the texture contour, adopt morphological operations to connect the broken edges to form a complete texture area, and separate the texture area from the background through a region growing algorithm or a threshold segmentation method; Texture feature extraction: extract the geometric features of the texture, including direction, spacing, depth, and shape; calculate the texture features of the texture; use Fourier transform or wavelet transform to analyze the frequency features of the texture; integrate the geometric features, texture features, and frequency features of the texture of a single region extracted from each high-definition image into the knife mark feature group, and integrate each knife mark feature group into the regional knife mark feature set corresponding to this single region.

3. The automatic monitoring method for the plane knife marks of an automotive accessory according to claim 1, wherein: The inputting this regional knife mark feature set, several other regional knife mark feature sets determined to have planar knife marks, and the functional attribute information of this automotive part into a depth analysis model includes: Determine the location of this regional knife mark feature set on the automotive part, and determine the target surface area of the automotive part based on this location, and determine each other regional knife mark feature set determined to have planar knife marks in the target surface area as the auxiliary regional knife mark feature sets; input this regional knife mark feature set, each of the auxiliary regional knife mark feature sets, and the functional attribute information of this automotive part into a depth analysis model, and the depth analysis model outputs a third knife mark probability; wherein, the functional attribute information includes the usage type and installation position of the automotive part; calculate the second knife mark probability according to the third knife mark probability.

4. The automatic monitoring method for the plane knife marks of an automotive part according to claim 3, characterized in that: Calculating the second knife mark probability based on the third knife mark probability includes: extracting the accessory features of the automotive accessory from any one of the high-definition images, and determining whether the area belongs to the inner or outer side of the automotive accessory based on the accessory features; if the area belongs to the inner side of the automotive accessory, configuring a first coefficient; if the area belongs to the outer side of the automotive accessory, configuring a second coefficient; wherein, the first coefficient is greater than the second coefficient; using the first coefficient or the second coefficient to optimize the third knife mark probability to obtain the second knife mark probability.

5. The automatic monitoring method for the flat knife marks of an automotive accessory according to claim 4, characterized in that: Determining the target surface area of the automotive accessory based on this area includes: determining the probability difference between the first knife mark probability and the first probability of the knife mark feature set in this area, and matching the surface radius according to the magnitude of the probability difference; wherein, the surface radius is positively correlated with the probability difference; based on this area, the target surface area is determined on the surface of the automotive accessory according to the surface radius.

6. An automatic monitoring system for the planar knife marks of automotive parts, the system comprising a processing device and a storage device, characterized in that: The computer code stored in the storage device is called and executed by the processing device to implement the following steps: under the illumination of a uniform light source, capturing high-definition images of the automotive accessory from multiple angles, and extracting knife mark features from each of the high-definition images to obtain multiple regional knife mark feature sets; wherein, each regional knife mark feature set contains multiple knife mark feature groups, and each knife mark feature group corresponds to a high-definition image at a certain angle; inputting each knife mark feature group in the regional knife mark feature set into a classifier respectively, and integrating the multiple knife mark probabilities output by the classifier to obtain the first knife mark probability corresponding to this regional knife mark feature set; if the first knife mark probability is higher than the first probability, it is determined that there are planar knife marks in the area of the automotive accessory corresponding to this regional knife mark feature set; if the first knife mark probability is lower than the first probability and higher than the second probability, input this regional knife mark feature set, several other regional knife mark feature sets determined to have planar knife marks, and the functional attribute information of this automotive accessory into a depth analysis model, and if the second knife mark probability output by the depth analysis model is higher than the first probability, it is determined that there are planar knife marks in the area of the automotive accessory corresponding to this regional knife mark feature set; performing label marking on the area of the automotive accessory determined to have planar knife marks.

7. An automatic monitoring system for the planar knife marks of automotive parts according to claim 6, characterized in that: Inputting this regional knife mark feature set, several other regional knife mark feature sets determined to have planar knife marks, and the functional attribute information of this automotive accessory into a depth analysis model includes: determining the location of this regional knife mark feature set on the automotive accessory, and determining the target surface area of the automotive accessory based on this location, and determining each other regional knife mark feature set determined to have planar knife marks corresponding to the target surface area as an auxiliary regional knife mark feature set; inputting this regional knife mark feature set, each of the auxiliary regional knife mark feature sets, and the functional attribute information of this automotive accessory into a depth analysis model, and the depth analysis model outputs a third knife mark probability; wherein, the functional attribute information includes the usage type and installation position of the automotive accessory; calculating the second knife mark probability based on the third knife mark probability.

8. An electronic device, comprising: At least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, characterized in that: the processor executes the computer program to implement the method according to any one of claims 1-5.

9. A computer storage medium storing a computer program, characterized in that: The computer program is executed by a processor to implement the method according to any one of claims 1-5.

10. A computer program product, characterized in that: The computer program product contains computer code, which, when executed by a processor of an electronic device, implements the method according to any one of claims 1-5.

Images