Safety shoe mixed line identification method, device and equipment and storage medium

Abstract

The invention provides a safety shoe mixed line identification method and device, equipment and a storage medium, and the method comprises the steps: obtaining the identification image information of a plurality of to-be-identified safety shoes, obtaining the corresponding contour information of the safety shoes, and determining a corresponding shooting angle; obtaining a preset safety shoe model library, wherein the preset safety shoe model library comprises a plurality of 3D models corresponding to preset safety shoes and preset safety shoe identifiers; obtaining 3D image information corresponding to each 3D model at the shooting angle; the 3D image information at least comprises a plane view, color information and depth information; respectively carrying out feature extraction on the color information and the depth information to obtain color feature information and depth feature information, and marking the color feature information and the depth feature information in corresponding plane views; and feature matching is carried out on the identification image information and the marked plane view, a preset safety shoe identifier corresponding to the successfully matched plane view is a target safety shoe identifier, the target safety shoe identifier is added to the corresponding identification image information, and mixed line identification of the plurality of safety shoes to be identified is completed.

Description

Technical Field

[0001] The present application relates to the technical field of industrial image processing, and in particular to a method, device, equipment and storage medium for identifying mixed lines of safety shoes. Background Art

[0002] In modern industrial production, safety shoes, as important personal protective equipment, are widely used in various working environments to protect workers' feet from injury. However, with the automation and diversification of production lines, the types and styles of safety shoes are also increasing, which brings new challenges to the identification and classification of safety shoes. Especially in mixed-line production environments, safety shoes of different models, colors and designs may coexist on the same production line. How to quickly and accurately identify and distinguish these safety shoes has become an urgent problem to be solved.

[0003] Traditional safety shoe identification methods mainly rely on manual inspection or simple visual recognition systems, which are not only inefficient but also easily affected by light, angle and environmental factors, resulting in high recognition error rates. Especially in large-scale production environments, manual recognition is not only time-consuming and labor-intensive, but also difficult to ensure consistency and accuracy. Summary of the invention

[0004] The present application provides a safety shoe mixed line identification method, device, equipment and storage medium, aiming to solve the problem that the traditional safety shoe identification method mainly relies on manual inspection or simple visual recognition system, which is not only inefficient, but also easily affected by light, angle and environmental factors, resulting in high recognition error rate. Especially in a large-scale production environment, manual identification is not only time-consuming and labor-intensive, but also difficult to ensure consistency and accuracy.

[0005] In a first aspect, the present application provides a method for identifying mixed lines of safety shoes, characterized in that the method comprises:

[0006] Obtaining identification image information of multiple safety shoes to be identified;

[0007] Acquire corresponding safety shoe contour information according to each of the recognition image information, so as to determine the shooting angle corresponding to the recognition image information according to the safety shoe contour information;

[0008] Obtain a preset safety shoe model library, wherein the safety shoe model library includes 3D models corresponding to a plurality of preset safety shoes and a preset safety shoe identifier corresponding to each of the 3D models;

[0009] Acquire 3D image information corresponding to each of the 3D models at the shooting angle; the 3D image information at least includes a plane view, color information and depth information;

[0010] Extracting features from the color information and the depth information respectively to obtain color feature information and depth feature information, and marking the color feature information and the depth feature information in the corresponding plan view;

[0011] The identified image information is feature matched with each of the annotated plan views, and the preset safety shoe logo corresponding to the successfully matched plan view is used as the target safety shoe logo. The target safety shoe logo is added to the corresponding identified image information to complete the mixed line recognition of multiple safety shoes to be identified.

[0012] In some embodiments, the safety shoe contour information includes geometric feature information corresponding to the safety shoes to be identified; the step of obtaining the corresponding safety shoe contour information according to each of the identified image information to determine the shooting angle corresponding to the identified image information according to the safety shoe contour information includes: extracting edge contour information corresponding to each of the identified image information according to a preset edge detection algorithm; obtaining the geometric feature information corresponding to the edge contour information; performing feature recognition according to the geometric feature information to obtain the position of key feature points in the edge contour information; the key feature points include at least the toe and the heel; and determining the shooting angle according to the position of the key feature points.

[0013] In some embodiments, obtaining the 3D image information corresponding to each of the 3D models at the shooting angle includes: determining a projection method for the 3D model according to the shooting angle; the projection method includes any one of an orthogonal projection or a perspective projection; rendering the 3D model according to the projection method and the projection angle based on ray tracing technology to obtain the plan view, color information, and depth information.

[0014] Exemplarily, before rendering the 3D model based on the projection mode and projection angle based on the ray tracing technology, it also includes: optimizing each of the 3D models based on detail level technology to reduce the number of polygons in each 3D model and improve the rendering speed of the 3D model.

[0015] In some embodiments, the feature extraction of the color information and the depth information respectively to obtain the color feature information and the depth feature information includes: determining the structural distribution information of the corresponding preset safety shoes according to each of the 3D models; generating a preset segmentation method according to the structural distribution information; respectively obtaining a color map corresponding to the color information and a depth map corresponding to the depth information; respectively segmenting the color map and the depth map according to the preset segmentation method to obtain multiple color local areas and multiple depth local areas; respectively calculating an overall color histogram corresponding to the color map and an overall depth histogram corresponding to the depth map; respectively calculating a local color histogram corresponding to each of the color local areas and a local depth histogram corresponding to each of the depth local areas; generating the color feature information according to each of the local color histograms and the overall color histogram, and generating the depth feature information according to each of the local depth histograms and the overall depth histogram.

[0016] In some embodiments, the feature matching of the recognized image information with each of the annotated plan views includes: determining a key matching area corresponding to each of the annotated plan views according to each of the annotated plan views; extracting a first area image corresponding to the recognized image information in the key matching area; extracting a second area image corresponding to the plan view in the key matching area; calculating an image matching degree between each of the first area images and the corresponding second area image; and determining a target matching degree based on a plurality of the image matching degrees, wherein the plan view corresponding to the target matching degree is the plan view that has successfully matched.

[0017] In some embodiments, the preset safety shoe identification includes a model identification, a size identification and a left-right foot identification; a plurality of the safety shoes to be identified are transferred in a preset production line, the production line includes a plurality of sorting ports, each sorting port corresponds to one of the model identifications; after the target safety shoe identification is added to the corresponding identification image information, it also includes: performing texture matching on the identification image information and the successfully matched plan view to obtain a texture matching result corresponding to the safety shoe; if it is determined that the texture of the safety shoe is complete according to the texture matching result, determining a target sorting port corresponding to the safety shoe to be identified among the plurality of sorting ports according to the model identification; controlling the production line to transfer the safety shoe to be identified to the target sorting port, and generating sorting information according to the size identification and the left-right foot identification, so as to determine a target shoe box corresponding to the safety shoe to be identified according to the sorting information, for sorting the safety shoe into the target shoe box.

[0018] In a second aspect, the present application provides a safety shoe mixed line identification device, the device comprising:

[0019] An image acquisition unit, used for acquiring identification image information of a plurality of safety shoes to be identified;

[0020] An angle acquisition unit, used for acquiring corresponding safety shoe contour information according to each of the recognition image information, so as to determine a shooting angle corresponding to the recognition image information according to the safety shoe contour information;

[0021] A model acquisition unit, used to acquire a preset safety shoe model library, wherein the safety shoe model library includes 3D models corresponding to a plurality of preset safety shoes and a preset safety shoe identifier corresponding to each of the 3D models;

[0022] A view acquisition unit, configured to acquire 3D image information corresponding to each of the 3D models at the shooting angle; the 3D image information at least includes a plane view, color information and depth information;

[0023] an information marking unit, configured to respectively extract features from the color information and the depth information, obtain color feature information and depth feature information, and mark the color feature information and the depth feature information in the corresponding plan view;

[0024] The recognition completion unit is used to perform feature matching between the recognition image information and each of the annotated plan views, and the preset safety shoe identifier corresponding to the successfully matched plan view is used as the target safety shoe identifier. The target safety shoe identifier is added to the corresponding recognition image information to complete the mixed-line recognition of multiple safety shoes to be identified.

[0025] In a third aspect, an embodiment of the present application provides a computer device, including a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and implement the safety shoe mixed line identification method provided in any embodiment of the present application when executing the computer program.

[0026] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements a safety shoe mixed line identification method as provided in any embodiment of the present application.

[0027] The present application provides a method, device, equipment and storage medium for identifying mixed lines of safety shoes. The method first obtains identification image information of multiple safety shoes to be identified, and then obtains the corresponding safety shoe contour information by processing each identification image information. The contour information is used to determine the shooting angle, providing a basis for subsequent accurate identification. This step is equivalent to preprocessing the image, so that subsequent identification can be more accurately matched according to the shooting angle.

[0028] Then, a preset safety shoe model library is obtained, which contains 3D models corresponding to multiple preset safety shoes and preset safety shoe logos corresponding to each 3D model. The model library is pre-established and covers various possible safety shoe styles and models, providing rich reference templates for identification.

[0029] Then, for each 3D model, its 3D image information under a certain shooting angle is obtained, which includes at least a plane view, color information, and depth information. By extracting features from the color information and depth information respectively, color feature information and depth feature information are obtained, and these feature information are annotated in the corresponding plane view. In this way, each plane view is not just a simple image, but contains rich color and depth features, which can more comprehensively describe the characteristics of the safety shoes.

[0030] Finally, the recognition image information of the safety shoes to be recognized is matched with each annotated plane view, and the plane view most similar to the image to be recognized is found by comparing the features of the image. Once the match is successful, the preset safety shoe logo corresponding to the plane view is determined as the target safety shoe logo and added to the corresponding recognition image information, thereby completing the mixed line recognition of multiple safety shoes to be recognized.

[0031] The provided method has at least the following beneficial effects:

[0032] 1. Improve recognition accuracy: Traditional manual inspection or simple visual recognition systems are easily affected by light, angle and environmental factors. This method obtains the safety shoe contour information to determine the shooting angle, and combines the multi-dimensional information of the 3D model for feature matching. It can more accurately identify safety shoes, reduce recognition errors caused by environmental and human factors, and improve recognition accuracy.

[0033] 2. Improve recognition efficiency: In a large-scale production environment, manual recognition is time-consuming and labor-intensive and it is difficult to ensure consistency and accuracy. This method uses a preset safety shoe model library and an automated feature extraction and matching algorithm to quickly identify multiple safety shoes to be identified, greatly improving recognition efficiency and meeting the needs of large-scale production.

[0034] 3. Realize mixed-line identification: It can effectively identify the mixed-line production of safety shoes of different styles and models, and can accurately distinguish various safety shoes, which provides convenience for the production management of enterprises, helps to realize the flexible production mode of multiple varieties and small batches, and improves the production flexibility and market competitiveness of enterprises.

[0035] 4. Strong scalability: This method is based on the principle of model library and feature matching. When a new style or model of safety shoes appears, you only need to add its 3D model and corresponding logo to the model library to identify the new safety shoes. It has good scalability and adaptability.

[0036] 5. Traceable results: By adding the target safety shoe logo to the corresponding identification image information, a clear record is provided for the identification results of each pair of safety shoes, which facilitates subsequent quality traceability and management, and helps companies monitor and optimize the production process.

[0037] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 is a schematic flow chart of the steps of a method for identifying mixed lines of safety shoes provided in one embodiment of the present application;

[0040] Figure 2 It is a schematic block diagram of the structure of a safety shoe mixed line identification device provided by an embodiment of the present application;

[0041] Figure 3 It is a schematic block diagram of the structure of a computer device provided in one embodiment of the present application.

[0042] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0044] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0045] It should be understood that, in order to facilitate the clear description of the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second", etc. are used to distinguish the same items or similar items with substantially the same functions and effects. Those skilled in the art can understand that the words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not necessarily limit the difference.

[0046] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0047] It should also be understood that the term “and / or” used in the specification and appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0048] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0049] In modern industrial production, safety shoes, as important personal protective equipment, are widely used in various working environments to protect workers' feet from injury. However, with the automation and diversification of production lines, the types and styles of safety shoes are also increasing, which brings new challenges to the identification and classification of safety shoes. Especially in mixed-line production environments, safety shoes of different models, colors and designs may coexist on the same production line. How to quickly and accurately identify and distinguish these safety shoes has become an urgent problem to be solved.

[0050] Traditional safety shoe identification methods mainly rely on manual inspection or simple visual recognition systems, which are not only inefficient but also easily affected by light, angle and environmental factors, resulting in high recognition error rates. Especially in large-scale production environments, manual recognition is not only time-consuming and labor-intensive, but also difficult to ensure consistency and accuracy.

[0051] To solve the above problems, the present application provides a safety shoe mixed line identification method, which can be carried out on a computer device, and the computer device can be deployed on a single server or a server cluster. It can also be deployed on a handheld terminal, a laptop computer, a wearable device or a robot, etc.

[0052] like Figure 1 As shown, the provided safety shoe mixed line identification method includes steps S101 to S106. The details are as follows:

[0053] Step S101: Obtain identification image information of a plurality of safety shoes to be identified.

[0054] Specifically, by capturing images of the safety shoes to be identified from the production line or other applicable environments. This can be obtained using a fixed industrial camera or a mobile device (such as a handheld scanner). The image should be high-definition and clearly show the various details of the safety shoes.

[0055] For example, on a safety shoe production line, multiple industrial cameras are installed, and when the safety shoes pass in front of the camera, the camera automatically captures images of the safety shoes. These images include different angles and parts of the shoes, such as the front, side, and sole.

[0056] This ensures the quality and diversity of the images and provides reliable basic data for subsequent accurate identification.

[0057] Step S102: Acquire corresponding safety shoe contour information according to each recognition image information, so as to determine the shooting angle corresponding to the recognition image information according to the safety shoe contour information.

[0058] Specifically, the image obtained in step S101 is analyzed by using image processing technology (such as Canny edge detection algorithm) to extract the outline of the safety shoes. Then, the orientation of the shoes in the image is analyzed based on the outline information to determine the camera angle when shooting.

[0059] For example, when analyzing an image, the computer device can identify whether the shoe is photographed from the front or side, or from a specific angle (such as 45 degrees). This process may involve using a machine learning algorithm to determine the relationship between the outline and a preset model to infer the angle. By accurately determining the shooting angle of each image, it helps to accurately match the plan view in the 3D model library in the subsequent steps.

[0060] Step S103: Obtain a preset safety shoe model library, wherein the safety shoe model library includes 3D models corresponding to a plurality of preset safety shoes and a preset safety shoe identifier corresponding to each 3D model.

[0061] Specifically, a database containing a variety of 3D models of safety shoes is established. Each 3D model has its own unique safety shoe logo, representing the specific type and model corresponding to the model. These 3D models can show the appearance of the shoes from multiple angles, including plane views, color information, depth information, etc.

[0062] For example, the model library contains 3D models of safety shoes of different brands and models. Each model has been carefully designed to clearly show the characteristics of the shoes from multiple perspectives. For example, the 3D model of a certain safety shoe not only shows the overall outline of the shoe, but also includes its unique color and texture details.

[0063] Having a detailed and comprehensive 3D model library provides standards for subsequent recognition and improves recognition accuracy.

[0064] Step S104: Obtain 3D image information corresponding to each 3D model at a shooting angle; the 3D image information at least includes a plane view, color information, and depth information.

[0065] Specifically, according to the shooting angle determined in step S102, a corresponding 3D model is selected from the 3D model library, and image information of the model at that angle is generated, including a plane view, color information, and depth information. This is a process of rendering a 3D model into a 2D image. If step S102 determines that the shooting angle is frontal, the system will render the front view of all preset safety shoes in the library, while retaining the color and depth information of the shoes.

[0066] This ensures that the 3D model view used for comparison is consistent with the captured image view, reducing matching errors caused by angle differences.

[0067] Step S105: extract features from the color information and the depth information respectively, obtain color feature information and depth feature information, and mark the color feature information and the depth feature information in the corresponding plan view.

[0068] Specifically, a deep learning algorithm (such as a convolutional neural network) is used to process the image information generated in step S104 to extract key features, including color features and depth features, and then these feature information are annotated on the corresponding 2D plane view to form a feature map.

[0069] For example, the computer equipment uses a convolutional neural network to extract the main color and icon of the safety shoes from the front view, and extracts the concave and convex features of the shoe surface, such as the texture of the sole, from the depth information. These features will be marked on the corresponding view to form a feature map that can be used for comparison.

[0070] By extracting and labeling key features, the subsequent matching process is simplified and the matching speed and accuracy are improved.

[0071] Step S106. Feature matching is performed between the recognized image information and each annotated plane view. The preset safety shoe identifier corresponding to the successfully matched plane view is used as the target safety shoe identifier. The target safety shoe identifier is added to the corresponding recognized image information to complete the mixed line recognition of multiple safety shoes to be identified.

[0072] Specifically, the recognition image information obtained in step S101 is compared with the feature map generated in step S105, and a feature matching algorithm (such as SIFT or SURF) is used to find the most similar model view. When the feature map with the highest matching degree is found, the preset safety shoe logo corresponding to the feature map is considered to be the logo of the shoe to be recognized.

[0073] Once the image features of the safety shoes are highly matched with a feature map in the model library, the system will automatically add the successfully matched model logo to the recognition image to complete the automatic recognition of the safety shoes. Fast and accurate safety shoe recognition is achieved, greatly improving the efficiency and safety of the production line. By reducing manual work, the recognition error caused by human error can also be reduced, while improving the consistency and reliability of the recognition process.

[0074] In summary, the provided safety shoe mixed line identification method not only solves the problems of low efficiency and error-proneness of traditional methods by combining modern image processing technology and deep learning algorithms, but also adapts to the trend of production automation and diversification, bringing significant improvements to the production management of safety shoes.

[0075] In some embodiments, the safety shoe contour information includes geometric feature information corresponding to the safety shoe to be identified; the corresponding safety shoe contour information is obtained according to each identified image information, so as to determine the shooting angle corresponding to the identified image information according to the safety shoe contour information, including: extracting the edge contour information corresponding to each identified image information according to a preset edge detection algorithm; obtaining the geometric feature information corresponding to the edge contour information; performing feature recognition according to the geometric feature information to obtain the position of key feature points in the edge contour information; the key feature points include at least the toe and the heel; and determining the shooting angle according to the position of the key feature points.

[0076] The edge contour information of the safety shoes is extracted from the recognition image by using a preset edge detection algorithm (such as Canny edge detection, Sobel operator, etc.). Based on the extracted edge contour information, the geometric feature information of the safety shoes is further obtained, such as the shape, size, angle, etc. of the contour. Through the feature recognition algorithm, the key feature point positions are identified from the geometric feature information, such as the positions of the toe and heel. Based on the position information of the key feature points, the shooting angle is determined to provide accurate viewing angle information for subsequent 3D model matching.

[0077] Suppose an industrial camera on a production line takes an image of a safety shoe. The computer first uses the Canny edge detection algorithm to extract the edge contour of the shoe, and then determines that this is a side view of the shoe by analyzing the geometric features of the contour (such as the positional relationship between the toe and the heel).

[0078] By extracting geometric feature information and key feature points, the shooting angle can be judged more accurately, thereby reducing recognition errors caused by angle errors. Extracting geometric features and key points is much simpler than directly processing the entire image, which improves processing speed and efficiency. Even when conditions such as light and background change, geometric features and key points can still maintain a high level of recognition stability.

[0079] In some embodiments, 3D image information corresponding to each 3D model at a shooting angle is obtained, including: determining a projection method for the 3D model according to the shooting angle; the projection method includes either an orthogonal projection or a perspective projection; rendering the 3D model according to the projection method and the projection angle based on ray tracing technology to obtain a plane view, color information, and depth information.

[0080] By selecting the appropriate projection method according to the shooting angle. Orthographic projection is suitable for front view and side view, and perspective projection is suitable for oblique view. Ray tracing technology is used to render the 3D model according to the determined projection method and projection angle to generate plane view, color information and depth information.

[0081] For example, the computer device selects a perspective projection mode according to the shooting angle determined in step S102 (eg, a 45-degree oblique view), and uses ray tracing technology to render each 3D model in the 3D model library to generate a plane view, a color map, and a depth map at corresponding angles.

[0082] Ray tracing technology can generate high-precision 3D image information that is highly similar to the real shot image, improving the accuracy of the match. By selecting the appropriate projection method, it is ensured that the generated image can cover a variety of shooting angles and adapt to complex working environments. Only the required viewing angle images are generated, avoiding unnecessary data processing and improving system efficiency.

[0083] Exemplarily, before rendering the 3D model according to the projection method and projection angle based on the ray tracing technology, it also includes: optimizing each 3D model based on the detail level technology to reduce the number of polygons in each 3D model and improve the rendering speed of the 3D model.

[0084] By optimizing each 3D model using the level of detail (LOD) technique before rendering, the number of polygons in the model is reduced. The optimized 3D model reduces the amount of calculation while maintaining the visual effect, thereby increasing the rendering speed.

[0085] For example, before rendering a 3D model, LOD technology is used to reduce the number of polygons in the model from 100,000 to 10,000 before rendering. The optimized model can generate the required plan view, color map, and depth map in a few seconds, while the unoptimized model may take tens of seconds or even longer.

[0086] By reducing the number of polygons, the rendering speed of the 3D model is significantly improved, thereby speeding up the entire recognition process. The optimized model reduces the demand for computing resources, allowing the system to run more tasks under limited hardware conditions. Despite the reduction in the number of polygons, the rendered image still maintains high visual quality and details through LOD technology, without affecting the final recognition effect.

[0087] In some embodiments, feature extraction is performed on color information and depth information respectively to obtain color feature information and depth feature information, including: determining the corresponding preset structural distribution information of safety shoes according to each 3D model; generating a preset segmentation method according to the structural distribution information; respectively obtaining a color map corresponding to the color information and a depth map corresponding to the depth information; respectively segmenting the color map and the depth map according to the preset segmentation method to obtain multiple color local areas and multiple depth local areas; respectively calculating an overall color histogram corresponding to the color map and an overall depth histogram corresponding to the depth map; respectively calculating a local color histogram corresponding to each color local area and a local depth histogram corresponding to each depth local area; generating color feature information according to each local color histogram and the overall color histogram, and generating depth feature information according to each local depth histogram and the overall depth histogram.

[0088] Generate a preset segmentation method based on the preset safety shoe structure distribution information of each 3D model. Obtain a color map and a depth map from the 3D image information of the 3D model respectively. Segment the color map and the depth map according to the preset segmentation method to generate multiple color local areas and multiple depth local areas. Calculate the overall color histogram of the color map and the overall depth histogram of the depth map. Calculate the local color histogram of each color local area and the local depth histogram of each depth local area. Generate color feature information based on each local color histogram and the overall color histogram. Generate depth feature information based on each local depth histogram and the overall depth histogram.

[0089] Assume that a safety shoe model in a 3D model library has clear structural distribution information, such as the specific locations of the toe, upper, sole, etc. The computer device generates a segmentation method based on the specific locations of the toe, upper, sole, etc., and divides the color map and depth map into multiple regions. Then, the color histogram and depth histogram of each region are calculated to generate detailed color feature information and depth feature information.

[0090] By combining local and overall histograms, the features of the safety shoes can be described more finely, and the accuracy of recognition can be improved. The comparison of local features and overall features can effectively reduce the mismatch problem caused by similar features. The method corresponding to this embodiment is particularly suitable for safety shoes with complex textures and color distributions, and can better capture these detailed features.

[0091] In some embodiments, feature matching is performed between the identified image information and each annotated plan view, including: determining a key matching area corresponding to each plan view based on each annotated plan view; extracting a first area image corresponding to the identified image information in the key matching area; extracting a second area image corresponding to the plan view in the key matching area; calculating an image matching degree between each first area image and the corresponding second area image; and determining a target matching degree based on multiple image matching degrees, wherein the plan view corresponding to the target matching degree is a successfully matched plan view.

[0092] According to the annotated plan view, determine the key matching area corresponding to each plan view. Extract the first area image corresponding to the key matching area from the recognition image. Extract the second area image corresponding to the key matching area from the plan view. Use an image matching algorithm (such as SIFT, SURF, etc.) to calculate the image matching degree between each first area image and the corresponding second area image. According to multiple image matching degrees, determine the highest matching degree, and the plan view corresponding to the matching degree is the successfully matched view.

[0093] For example, by marking key matching areas such as the toe, upper and sole in the plane view, the images of these areas are extracted from the recognition image, and then the matching degree is calculated with the corresponding areas in the plane view. Finally, the plane view with the highest matching degree is selected as the successfully matched view.

[0094] By limiting the key matching area, the image area that needs to be matched is reduced and the matching speed is improved. The extraction and matching of the key matching area can more accurately capture and compare important features and improve the accuracy of matching. Matching of specific areas is more reliable than matching the entire image, and can effectively reduce misjudgments caused by background interference.

[0095] In some embodiments, a preset safety shoe identification includes a model identification, a size identification, and a left-right foot identification; a plurality of safety shoes to be identified are transferred in a preset production line, and the production line includes a plurality of sorting ports, each sorting port corresponding to a model identification; after adding the target safety shoe identification to the corresponding recognition image information, it also includes: performing texture matching based on the recognition image information and the successfully matched plane view to obtain the texture matching result corresponding to the safety shoe; if it is determined according to the texture matching result that the texture of the safety shoe is complete, determining the target sorting port corresponding to the safety shoe to be identified in the plurality of sorting ports according to the model identification; controlling the production line to transfer the safety shoes to be identified to the target sorting port, and generating sorting information according to the size identification and the left-right foot identification, so as to determine the target shoe box corresponding to the safety shoes to be identified according to the sorting information, for sorting the safety shoes into the target shoe box.

[0096] The identification of each safety shoe includes model, size, and left and right foot information. Further texture matching is performed between the recognized image information and the successfully matched plane view to obtain the texture matching result corresponding to the safety shoe. If the texture matching result shows that the texture of the safety shoe is complete, the target sorting port of the safety shoe to be identified on the production line is determined according to the model identification. Detailed sorting information is generated according to the size identification and left and right foot identification to control the production line to sort the safety shoes into the corresponding target shoe boxes.

[0097] Assume that after the safety shoes on the production line are identified, their model is determined to be "HS1001", size is "42", and they are left-foot shoes. Based on this information, the system determines that the target sorting port for the shoes is "D sorting port" and sends the sorting information (model, size, left and right feet) to the production line control unit to sort the shoes into the corresponding shoe boxes.

[0098] The automated sorting of safety shoes is realized, which reduces manual operations and improves the efficiency of the production line. Through texture matching and the generation of detailed identification information, it ensures that each pair of shoes is correctly sorted into the corresponding shoe box, avoiding misclassification and omission. The automated sorting system can monitor and manage the production process in real time, improving the overall production management level and quality control capabilities.

[0099] See also Figure 2 , Figure 2 1 is a schematic diagram of the structure of a safety shoe mixed line identification device 200 for an electric power system provided in an embodiment of the present application. The safety shoe mixed line identification device 200 for an electric power system is used to execute the steps of the safety shoe mixed line identification method shown in any embodiment of the present application. The safety shoe mixed line identification device 200 for an electric power system may be a single server or a server cluster, or the safety shoe mixed line identification device 200 for an electric power system may be a terminal, which may be a handheld terminal, a laptop computer, a wearable device, a robot, etc.

[0100] like Figure 2 As shown, the safety shoe mixed line identification device 200 includes:

[0101] An image acquisition unit 201 is used to acquire identification image information of a plurality of safety shoes to be identified;

[0102] An angle acquisition unit 202, used for acquiring corresponding safety shoe contour information according to each of the recognition image information, so as to determine the shooting angle corresponding to the recognition image information according to the safety shoe contour information;

[0103] A model acquisition unit 203 is used to acquire a preset safety shoe model library, wherein the safety shoe model library includes 3D models corresponding to a plurality of preset safety shoes and a preset safety shoe identifier corresponding to each of the 3D models;

[0104] A view acquisition unit 204 is used to acquire 3D image information corresponding to each of the 3D models at the shooting angle; the 3D image information at least includes a plane view, color information and depth information;

[0105] An information labeling unit 205 is used to perform feature extraction on the color information and the depth information respectively, obtain color feature information and depth feature information, and label the color feature information and the depth feature information in the corresponding plan view;

[0106] The recognition completion unit 206 is used to perform feature matching between the recognition image information and each of the annotated plan views, and the preset safety shoe identifier corresponding to the successfully matched plan view is used as the target safety shoe identifier, and the target safety shoe identifier is added to the corresponding recognition image information to complete the mixed line recognition of multiple safety shoes to be identified.

[0107] It should be noted that technical personnel in the relevant field can clearly understand that, for the convenience and conciseness of description, the specific working process of the safety shoe mixed line identification device 200 and each module described above can refer to the corresponding process in the safety shoe mixed line identification method embodiment described in any embodiment of the present application, and will not be repeated here.

[0108] Figure 1 The corresponding safety shoe mixed line identification method can be implemented in the form of a computer program, which can be used in Figure 2 Run on the device shown.

[0109] See also Figure 3 , Figure 3 1 is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application. The computer device includes a processor, a memory and a network interface connected via a device bus, wherein the memory may include a storage medium and an internal memory.

[0110] The storage medium can store operating devices and computer programs. The computer program includes program instructions, which, when executed, can cause the processor to execute Figure 2 Corresponding to any one of the safety shoe mixed line identification methods.

[0111] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0112] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any one of the safety shoe mixed line identification methods.

[0113] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the terminal to which the scheme of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0114] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0115] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0116] Obtaining identification image information of multiple safety shoes to be identified;

[0117] Acquire corresponding safety shoe contour information according to each of the recognition image information, so as to determine the shooting angle corresponding to the recognition image information according to the safety shoe contour information;

[0118] Obtain a preset safety shoe model library, wherein the safety shoe model library includes 3D models corresponding to a plurality of preset safety shoes and a preset safety shoe identifier corresponding to each of the 3D models;

[0119] Acquire 3D image information corresponding to each of the 3D models at the shooting angle; the 3D image information at least includes a plane view, color information and depth information;

[0120] Extracting features from the color information and the depth information respectively to obtain color feature information and depth feature information, and marking the color feature information and the depth feature information in the corresponding plan view;

[0121] The recognition image information is feature matched with each of the annotated plan views, and the preset safety shoe logo corresponding to the successfully matched plan view is used as the target safety shoe logo. The target safety shoe logo is added to the corresponding recognition image information to complete the mixed line recognition of multiple safety shoes to be identified.

[0122] In some embodiments, the safety shoe contour information includes geometric feature information corresponding to the safety shoes to be identified; the step of obtaining the corresponding safety shoe contour information according to each of the identified image information to determine the shooting angle corresponding to the identified image information according to the safety shoe contour information includes: extracting edge contour information corresponding to each of the identified image information according to a preset edge detection algorithm; obtaining the geometric feature information corresponding to the edge contour information; performing feature recognition according to the geometric feature information to obtain the position of key feature points in the edge contour information; the key feature points include at least the toe and the heel; and determining the shooting angle according to the position of the key feature points.

[0123] In some embodiments, obtaining the 3D image information corresponding to each of the 3D models at the shooting angle includes: determining a projection method for the 3D model according to the shooting angle; the projection method includes any one of an orthogonal projection or a perspective projection; rendering the 3D model according to the projection method and the projection angle based on ray tracing technology to obtain the plan view, color information, and depth information.

[0124] Exemplarily, before rendering the 3D model based on the projection mode and projection angle based on the ray tracing technology, it also includes: optimizing each of the 3D models based on detail level technology to reduce the number of polygons in each 3D model and improve the rendering speed of the 3D model.

[0125] In some embodiments, the feature extraction of the color information and the depth information respectively to obtain the color feature information and the depth feature information includes: determining the structural distribution information of the corresponding preset safety shoes according to each of the 3D models; generating a preset segmentation method according to the structural distribution information; respectively obtaining a color map corresponding to the color information and a depth map corresponding to the depth information; respectively segmenting the color map and the depth map according to the preset segmentation method to obtain multiple color local areas and multiple depth local areas; respectively calculating an overall color histogram corresponding to the color map and an overall depth histogram corresponding to the depth map; respectively calculating a local color histogram corresponding to each of the color local areas and a local depth histogram corresponding to each of the depth local areas; generating the color feature information according to each of the local color histograms and the overall color histogram, and generating the depth feature information according to each of the local depth histograms and the overall depth histogram.

[0126] In some embodiments, the feature matching of the recognized image information with each of the annotated plan views includes: determining a key matching area corresponding to each of the annotated plan views according to each of the annotated plan views; extracting a first area image corresponding to the recognized image information in the key matching area; extracting a second area image corresponding to the plan view in the key matching area; calculating an image matching degree between each of the first area images and the corresponding second area image; and determining a target matching degree based on a plurality of the image matching degrees, wherein the plan view corresponding to the target matching degree is the plan view that has successfully matched.

[0127] In some embodiments, the preset safety shoe identification includes a model identification, a size identification and a left-right foot identification; a plurality of the safety shoes to be identified are transferred in a preset production line, the production line includes a plurality of sorting ports, each sorting port corresponds to one of the model identifications; after the target safety shoe identification is added to the corresponding identification image information, it also includes: performing texture matching on the identification image information and the successfully matched plan view to obtain a texture matching result corresponding to the safety shoe; if it is determined that the texture of the safety shoe is complete according to the texture matching result, determining a target sorting port corresponding to the safety shoe to be identified among the plurality of sorting ports according to the model identification; controlling the production line to transfer the safety shoe to be identified to the target sorting port, and generating sorting information according to the size identification and the left-right foot identification, so as to determine a target shoe box corresponding to the safety shoe to be identified according to the sorting information, for sorting the safety shoe into the target shoe box.

[0128] The present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the steps of the safety shoe mixed line identification method provided in any embodiment corresponding to the present application.

[0129] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.

[0130] It should be noted that technical personnel in the relevant field can clearly understand that, for the convenience and conciseness of description, the specific structure of the ultra-thin high-temperature colored glaze glass with high flatness described above can refer to the corresponding process in the embodiments of the ultra-thin high-temperature colored glaze glass processing method with high flatness described in the above embodiments, and will not be repeated here.

[0131] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for identifying mixed lines of safety shoes, characterized in that: The method comprises: Obtaining identification image information of multiple safety shoes to be identified; Acquire corresponding safety shoe contour information according to each of the recognition image information, so as to determine the shooting angle corresponding to the recognition image information according to the safety shoe contour information; Obtain a preset safety shoe model library, wherein the safety shoe model library includes 3D models corresponding to a plurality of preset safety shoes and a preset safety shoe identifier corresponding to each of the 3D models; Acquire 3D image information corresponding to each of the 3D models at the shooting angle; the 3D image information at least includes a plane view, color information and depth information; Extracting features from the color information and the depth information respectively to obtain color feature information and depth feature information, and marking the color feature information and the depth feature information in the corresponding plan view; The recognition image information is feature matched with each of the annotated plan views, and the preset safety shoe logo corresponding to the successfully matched plan view is used as the target safety shoe logo. The target safety shoe logo is added to the corresponding recognition image information to complete the mixed line recognition of multiple safety shoes to be identified.

2. The method according to claim 1, characterized in that The safety shoe profile information includes geometric feature information corresponding to the safety shoe to be identified; The step of acquiring corresponding safety shoe contour information according to each of the recognition image information, so as to determine a shooting angle corresponding to the recognition image information according to the safety shoe contour information, comprises: Extract edge contour information corresponding to each of the identified image information according to a preset edge detection algorithm; Acquire the geometric feature information corresponding to the edge contour information; Performing feature recognition according to the geometric feature information to obtain key feature point positions in the edge contour information; the key feature points include at least a toe and a heel; The shooting angle is determined according to the position of the key feature point.

3. The method according to claim 1, characterized in that The obtaining of 3D image information corresponding to each of the 3D models at the shooting angle includes: Determining a projection method for the 3D model according to the shooting angle; the projection method includes any one of orthogonal projection or perspective projection; The 3D model is rendered according to the projection mode and projection angle based on ray tracing technology to obtain the plane view, color information and depth information.

4. The method according to claim 3, characterized in that Before rendering the 3D model based on the ray tracing technology according to the projection mode and projection angle, the method further includes: Each of the 3D models is optimized based on detail level technology to reduce the number of polygons in each 3D model and improve the rendering speed of the 3D model.

5. The method according to claim 1, characterized in that The extracting features of the color information and the depth information respectively to obtain the color feature information and the depth feature information includes: Determine the structural distribution information of the corresponding preset safety shoes according to each of the 3D models; generating a preset segmentation method according to the structural distribution information; Respectively obtaining a color map corresponding to the color information and a depth map corresponding to the depth information; Segment the color map and the depth map respectively according to the preset segmentation method to obtain a plurality of color local areas and a plurality of depth local areas; Respectively calculating an overall color histogram corresponding to the color map and an overall depth histogram corresponding to the depth map; Respectively calculating a local color histogram corresponding to each of the color local regions and a local depth histogram corresponding to each of the depth local regions; The color feature information is generated according to each of the local color histograms and the overall color histogram, and the depth feature information is generated according to each of the local depth histograms and the overall depth histogram.

6. The method according to claim 1, characterized in that The performing feature matching between the recognized image information and each of the annotated plan views includes: Determine a key matching area corresponding to each of the plan views according to each of the marked plan views; Extracting the first area image corresponding to the key matching area of ​​the recognized image information; Extracting a second area image corresponding to the key matching area of ​​the plan view; Calculating the image matching degree between each first region image and the corresponding second region image; A target matching degree is determined according to the plurality of image matching degrees, and the plane view corresponding to the target matching degree is the plane view that has been matched successfully.

7. The method according to claim 1, characterized in that The preset safety shoe identification includes a model identification, a size identification, and a left and right foot identification; a plurality of the safety shoes to be identified are transferred in a preset production line, the production line includes a plurality of sorting ports, each sorting port corresponds to one of the model identifications; After adding the target safety shoe logo to the corresponding recognition image information, the method further includes: Performing texture matching on the recognized image information and the successfully matched plan view to obtain a texture matching result corresponding to the safety shoe; If it is determined according to the texture matching result that the texture of the safety shoe is complete, the target sorting port corresponding to the safety shoe to be identified among the multiple sorting ports is determined according to the model identifier; The production line is controlled to transfer the safety shoes to be identified to the target sorting port, and sorting information is generated according to the size identification and the left and right foot identifications, so as to determine the target shoe box corresponding to the safety shoes to be identified according to the sorting information, so as to sort the safety shoes into the target shoe box.

8. A safety shoe mixed line identification device, characterized in that: The device comprises: An image acquisition unit, used for acquiring identification image information of a plurality of safety shoes to be identified; An angle acquisition unit, used for acquiring corresponding safety shoe contour information according to each of the recognition image information, so as to determine a shooting angle corresponding to the recognition image information according to the safety shoe contour information; A model acquisition unit, used to acquire a preset safety shoe model library, wherein the safety shoe model library includes 3D models corresponding to a plurality of preset safety shoes and a preset safety shoe identifier corresponding to each of the 3D models; A view acquisition unit, configured to acquire 3D image information corresponding to each of the 3D models at the shooting angle; the 3D image information at least includes a plane view, color information and depth information; an information marking unit, configured to respectively extract features from the color information and the depth information, obtain color feature information and depth feature information, and mark the color feature information and the depth feature information in the corresponding plan view; The recognition completion unit is used to perform feature matching between the recognition image information and each of the annotated plan views, and the preset safety shoe identifier corresponding to the successfully matched plan view is used as the target safety shoe identifier. The target safety shoe identifier is added to the corresponding recognition image information to complete the mixed-line recognition of multiple safety shoes to be identified.

9. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and implement the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the method according to any one of claims 1 to 7.

Images