Standard detection method, device and equipment for store materials and storage medium

Abstract

The embodiment of the invention provides a standard detection method and device for store materials, equipment and a storage medium, and belongs to the technical field of image processing. The standard detection method comprises the following steps: determining and cutting at least one first material image from at least one store scene image of a to-be-detected store; based on a preset material classification model, comparing the feature data of each cut first material image with feature data in a preset feature library, and determining the material category of each first material image according to a feature comparison result; and based on the material category of each first material image, according to a preset material specification standard, judging whether the material category of the store accords with the specification. According to the standard detection method provided by the embodiment of the invention, the preset material classification model is adopted, the material category to which the cut material image belongs can be accurately identified, and the accuracy of standard detection of store materials is ensured.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing technology, and in particular to a method, device, equipment and storage medium for standardizing store material detection. Background Art

[0002] Smart inspection of store materials is the process of using modern technology in the retail industry to automatically monitor and inspect materials used in stores, such as hanging flags, posters, floor stickers, etc.

[0003] In traditional store material standardization inspection, it is usually necessary to extract text information from material images for recognition to determine whether the materials are standard. However, this method is prone to recognition errors when faced with poor image quality, such as insufficient resolution, inaccurate focus, or blurred text, which in turn affects the accurate assessment of the standardization of store materials. Summary of the invention

[0004] The purpose of the embodiments of the present disclosure is to provide a method, device, equipment and storage medium for standard detection of store materials, aiming to solve the problem of inaccurate standard detection of store materials in traditional methods.

[0005] In order to achieve the above-mentioned objectives, in a first aspect, an embodiment of the present disclosure provides a standardized detection method for store materials, the standardized detection method comprising: determining and cropping at least one first material image from at least one store scene image of the store to be inspected; based on a preset material classification model, comparing feature data of each cropped first material image with feature data in a preset feature library, and determining the material category of each first material image according to the feature comparison result; based on the material category of each first material image, judging whether the material category of the store meets the specification according to the preset material specification standard.

[0006] In some embodiments, determining and cropping at least one first material image from at least one store scene image of a store to be inspected includes: acquiring at least one store scene image; determining the coordinate information of each material in at least one store scene image based on a preset material detection model; and cropping at least one first material image from at least one store scene image based on the coordinate information of each material, wherein the first material image is a single material image with marked coordinate information.

[0007] In some embodiments, the feature data in the preset feature library includes feature data of each second material image in at least two angular directions, wherein the second material image is a single material image with a marked material category.

[0008] In some embodiments, the feature data in the preset feature library includes feature data of each second material image in directions of 0°, 90°, 180°, and 270°.

[0009] In some embodiments, the feature data of each cropped first material image is compared with the feature data in a preset feature library, including: extracting a multidimensional feature vector of the cropped first material image; based on the multidimensional feature vector of the first material image, determining the cosine similarity between the multidimensional feature vector of the first material image and the multidimensional feature vector of the compared second material image in at least two angular directions; and using the cosine similarity as the feature comparison result.

[0010] In some embodiments, the material category of each first material image is determined based on the feature comparison result, including: when at least one cosine similarity in the feature comparison result is greater than a first threshold, determining that the material category of the first material image is the material category of the second material image being compared; when all cosine similarities in the feature comparison result are not greater than the first threshold, comparing the feature data of the first material image with the feature data of the next second material image in at least two angular directions, and determining the material category of the first material image again based on the feature comparison result.

[0011] In some embodiments, based on the material category of each first material image and according to preset material specification standards, determining whether the material category of the store meets the specifications includes: determining all material categories of the store and the material quantities corresponding to each material category based on the material category of each first material image; determining whether all material categories of the store and the material quantities corresponding to each material category meet the specifications according to preset material specification standards, wherein the preset material specification standards include at least one material category and the material quantity corresponding to each material category.

[0012] In a second aspect, an embodiment of the present disclosure provides a standard detection device for store materials, the standard detection device comprising: a first processing unit, for determining and cropping at least one first material image from at least one store scene image of the store to be detected; a second processing unit, for comparing feature data of each cropped first material image with feature data in a preset feature library based on a preset material classification model, and determining the material category of each first material image according to the feature comparison result; a judgment unit, for judging whether the material category of the store meets the specifications based on the material category of each first material image and according to a preset material specification standard.

[0013] In a third aspect, an embodiment of the present disclosure provides an electronic device, comprising: a memory; and a processor, wherein the processor is configured to execute the standardized detection method for store materials provided in the first aspect or any embodiment of the first aspect.

[0014] In a fourth aspect, an embodiment of the present disclosure provides a machine-readable storage medium having instructions stored thereon, the instructions being used to enable a machine to execute the standardized inspection method for store materials provided in the first aspect or any one of the embodiments of the first aspect.

[0015] Through the above technical solution, the standardized detection method provided by the embodiment of the present disclosure adopts a preset material classification model, which can accurately identify the material category to which the cropped material image belongs, thereby ensuring the accuracy of standardized detection of store materials.

[0016] Other features and advantages of the embodiments of the present disclosure will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present disclosure but do not constitute a limitation on the embodiments of the present disclosure. In the accompanying drawings:

[0018] Figure 1 It is a flow chart provided according to the first embodiment of the standardized detection method for store materials disclosed in the present invention;

[0019] Figure 2 It is a flow chart provided according to the second embodiment of the standardized detection method for store materials disclosed in the present invention;

[0020] Figure 3 It is a flow chart provided according to the third embodiment of the standardized detection method for store materials disclosed in the present invention;

[0021] Figure 4 It is a flow chart provided according to the fourth embodiment of the standardized detection method for store materials disclosed in the present invention;

[0022] Figure 5 It is a structural schematic diagram of a standardized detection device for store materials provided according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] The specific implementation of the embodiment of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present disclosure, and is not used to limit the embodiment of the present disclosure.

[0024] Smart inspection of store materials is a vital link in modern retail industry. It uses high-tech means to automatically monitor and inspect the atmosphere facilities and environment such as hanging flags, posters and floor stickers in the store.

[0025] In the prior art, a target detection model is usually used to detect specific materials in store scene images, and then text is extracted from the material images and compared with preset text to determine whether the target material exists in the store.

[0026] The above technical solution mainly compares the text extracted from the material image to determine whether the material is the target material. However, text detection requires high image clarity. Blurred images can easily lead to errors in text recognition, which in turn affects the comparison results and causes errors in material detection. In addition, this method is not suitable for materials with less text information.

[0027] At the same time, the applicant of this application also found that: 1) Since the monitoring equipment is mainly used for in-store security monitoring and is located at a high position, when using the monitoring equipment to obtain material images, there may be problems such as unclear images, image rotation, distortion and occlusion. 2) Since most of the acquired material images are not taken from the front, the rotation of the material image will affect the material classification and identification process.

[0028] Therefore, the traditional standardized testing methods of stores have certain limitations, resulting in inaccurate standardized testing of store materials.

[0029] Based on this, the present disclosure provides a standardized detection method for store materials, referring to Figure 1 As shown, Figure 1 It is a flow chart provided according to the first embodiment of the standardized testing method for store materials disclosed in the present invention.

[0030] like Figure 1 As shown, the standard detection method for store materials includes steps S101 to S103.

[0031] Step S101: determine and crop at least one first material image from at least one store scene image of the store to be detected.

[0032] In the disclosed embodiment, at least one store scene image of the store to be inspected can be obtained by at least one of the following methods: real-time shooting using a surveillance camera installed in the store, or taking photos at the store site using a smart device carried by sales staff or patrol personnel.

[0033] For example, store scene image data is obtained with the help of store monitoring acquisition equipment. The equipment needs to have a high-resolution camera to obtain high-definition video of the store. The camera should be able to work under different lighting conditions and can capture clear and complete images by adjusting the focal length and angle. After obtaining the video data, the store scene image data at the target time point is segmented according to the specific inspection time.

[0034] By accurately cropping specific material images from store scene images, the accuracy and processing speed of material recognition can be improved. Specifically, high-resolution cameras can be used to obtain clear store videos under different lighting conditions, and then image data can be segmented at specific time points to eliminate background interference and focus on the material itself, thereby optimizing recognition results and improving system robustness, ensuring that standardized inspections can still be accurately performed in a changing store environment.

[0035] Step S102, based on a preset material classification model, compare the feature data of each cropped first material image with the feature data in a preset feature library, and determine the material category of each first material image according to the feature comparison result.

[0036] In some embodiments, the feature data in the preset feature library may include feature data of each second material image in at least two angular directions, wherein the second material image may be a single material image of a labeled material category. The at least two angular directions include but are not limited to 45°, 60°, 90°, 120°, 180°, 270°, etc. Preferably, the feature data in the preset feature library may include feature data of each second material image in the directions of 0°, 90°, 180° and 270°. Such angular direction data can cover most of the rotation features, thereby improving the generalization ability of the model for complex scenes.

[0037] Specifically, the material cut in step S101 is input into the material classification model to extract features, and compared with the standard material feature data in the feature library to achieve automatic classification of the material image.

[0038] The disclosed embodiment is based on a preset material classification model, which can compare the feature data of each first material image with the feature data in a preset feature library. The feature data in the preset feature library contains the features of a single material image with annotated material categories in multiple angles, such as feature data in the directions of 0°, 90°, 180°, and 270°. The material classification model takes into account the tilt problem of the image that may be caused by the shooting angle, thereby improving the accuracy and robustness of recognition.

[0039] Step S103, based on the material category of each first material image and according to the preset material specification standard, it is determined whether the material category of the store meets the specification.

[0040] In the embodiment of the present disclosure, the main focus is on standard inspection of material categories and material quantities of store materials. Specifically, the embodiment of the present disclosure can automatically verify whether the store displays the correct category of materials and ensure that the placement quantity of these materials strictly complies with established specifications and standards.

[0041] In a specific implementation, the complete process of the first embodiment of the present disclosure may be: first, obtain material images with the help of store surveillance cameras, and then perform target detection of materials to obtain images and positions of individual materials. Perform feature extraction on individual material images, compare with data in the standard material feature database, and determine the material classification. Compare the material detection and recognition results with the standard material placement requirements, determine whether the material placement meets the requirements, and return the results.

[0042] The standard detection method for store materials provided in the embodiment of the present disclosure first cuts out specific material images from the store scene image, then uses an advanced material classification model to extract and compare features of these images, and finally determines whether the materials are compliant according to preset standard specifications. This method improves the accuracy and real-time performance of material management, reduces the consumption of human resources, and greatly improves work efficiency. In addition, through precise image recognition and classification, this method ensures that the placement of store materials meets the requirements of brand image and marketing activities, thereby enhancing customer experience and improving the economic benefits of the store.

[0043] Based on the first embodiment, the present disclosure provides a standardized detection method for store materials. Figure 2 As shown, Figure 2 It is a flow chart provided according to Example 2 of the standardized testing method for store materials disclosed in the present invention.

[0044] In a feasible implementation, at least one first material image is determined and cropped from at least one store scene image of the store to be detected, including steps S201 to S203.

[0045] Step S201, obtaining at least one store scene image.

[0046] In the disclosed embodiment, store scene image data is mainly acquired by means of store monitoring and acquisition equipment.

[0047] Step S202: determining coordinate information of each material in at least one store scene image based on a preset material detection model.

[0048] The preset material detection model can be YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector) or Faster R-CNN (Faster Region-based Convolutional Neural Network) based on deep learning. These models can identify the materials in the image and provide their bounding box coordinates.

[0049] In the disclosed embodiment, a real data set consisting of material images can be used to annotate the materials in the images (including rotation angles and coordinate information) and train a material detection model. The material detection model is applied to infer the collected images to obtain the location information of all materials in the collected images. The model can handle materials of different sizes and shapes, and is robust to the direction, rotation, and partial occlusion of the materials. The material detection standard is set as whether the material is detected or not, that is, whether the store has placed the material.

[0050] Step S203: based on the coordinate information of each material, at least one first material image is cropped from at least one store scene image.

[0051] The first material image is a single material image with marked coordinate information.

[0052] If a material is detected by the material detection model, the material is cropped out of the store scene based on the material's coordinate information in the image; if a material is not detected by the material detection model, this result is returned and a notification mechanism is automatically triggered to remind store managers or sales staff to replenish or adjust the placement of materials in a timely manner, ensuring that the store's material display always complies with brand specifications and marketing strategies.

[0053] The material detection model provided by the embodiment of the present disclosure takes into account the rotation of the image, and when fine-tuning the material detection model, the training data set includes material images actually taken at different angles and lighting conditions, which can improve the accuracy of the target detection model in detecting materials under complex conditions to a certain extent.

[0054] Based on the first embodiment, the present disclosure provides a standardized detection method for store materials. Figure 3 As shown, Figure 3 It is a flow chart provided according to Example 3 of the standardized testing method for store materials disclosed in the present invention.

[0055] In a feasible implementation manner, comparing the feature data of each cropped first material image with the feature data in a preset feature library includes steps S301 to S303.

[0056] Step S301, extracting a multi-dimensional feature vector of the cropped first material image.

[0057] In the embodiments of the present disclosure, deep learning models, such as convolutional neural networks (CNN), deep convolutional neural networks (VGG), etc., can be used to extract features of a single material image. The feature data may include low-level features such as color, edge, texture, etc., and may also include deep abstract features of the image.

[0058] In image processing, feature vectors can be used to represent images, and image compression can be achieved by retaining the most important features, thereby reducing storage requirements. A multidimensional feature vector can represent multiple features of a data point, such as color, size, and shape. Given that materials usually contain a variety of features, the disclosed embodiment can extract a multidimensional feature vector from the cropped first material image.

[0059] For example, feature extraction is performed on the cropped first material image to obtain a feature vector with a dimension of 1*1000.

[0060] The disclosed embodiment can comprehensively capture multiple key attributes of the image by extracting multi-dimensional feature vectors of the cropped material image. Such high-dimensional data representation not only increases the richness of information, improves the accuracy and robustness of recognition, but also enhances the model's generalization ability for complex scenarios, thereby achieving more efficient and reliable material detection and classification in intelligent inspections of store materials.

[0061] Step S302: based on the multidimensional feature vector of the first material image, determine the cosine similarity between the multidimensional feature vector of the first material image and the multidimensional feature vector of the second material image to be compared in at least two angular directions.

[0062] It should be noted that cosine similarity is an indicator to measure the directional similarity of two vectors in multidimensional space. The similarity of the two vectors can be evaluated by calculating the cosine value of the angle between the two vectors. The closer the cosine value is to 1, the more similar the two vectors are. In addition to cosine similarity, there are other similarity or distance measurement methods, such as Euclidean distance, Manhattan distance, Jaccard similarity coefficient, etc. Each method has its specific application scenarios and advantages and disadvantages. When choosing an appropriate measurement method, it is necessary to consider the characteristics of the data and the specific needs of the application.

[0063] In some embodiments, the material category of each first material image is determined based on the feature comparison result, including: when at least one cosine similarity in the feature comparison result is greater than a first threshold, determining that the material category of the first material image is the material category of the second material image being compared; when all cosine similarities in the feature comparison result are not greater than the first threshold, comparing the feature data of the first material image with the feature data of the next second material image in at least two angular directions, and determining the material category of the first material image again based on the feature comparison result.

[0064] For example, after obtaining a 1*1000-dimensional feature vector of a single material image, the cosine similarity is calculated between it and the standard feature vector in the feature library (also a 1*1000-dimensional vector). If the similarity exceeds 0.5, it is considered that the material image and the image in the feature library are the same material image. That is, if the feature data of the image of the material to be tested matches the target material image in any of the four directions of 0°, 90°, 180° and 270°, the image of the material to be tested is considered to be the target material image. If the similarity does not exceed 0.5, it is compared with the next image again.

[0065] Step S303: taking the cosine similarity as the feature comparison result.

[0066] When classifying the cropped material images using a preset material classification model, the disclosed embodiment determines whether they are the same image by comparing the cosine similarity between a single material image and the material image rotated in at least two angular directions (for example, 0°, 90°, 180°, and 270°), thereby avoiding classification errors caused by rotation problems and improving the accuracy of material recognition.

[0067] Based on the first embodiment, the present disclosure provides a standardized detection method for store materials. Figure 4 As shown, Figure 4 It is a flow chart provided according to the fourth embodiment of the standardized testing method for store materials disclosed in the present invention.

[0068] In a feasible implementation, based on the material category of each first material image and according to a preset material specification standard, it is determined whether the material category of the store meets the specification, including steps S401 to S402.

[0069] Step S401: Based on the material category of each first material image, determine all material categories of the store and the material quantity corresponding to each material category.

[0070] Step S402: According to the preset material specification standard, determine whether all material categories of the store and the material quantity corresponding to each material category meet the specification.

[0071] The preset material specification standards may include at least one material category and the material quantity corresponding to each material category. In some special scenarios, the preset material specification standards may also include the display time of the material, the size of the material, etc.

[0072] Specifically, take the preset material specification standard of placing a 30% off billboard as an example. If the store places exactly one 30% off billboard, it meets the preset material specification standard; if the store places more or less than one 30% off billboard, for example, two or none, it is considered that the quantity does not match; if the store places other types of materials other than 30% off billboards, such as 50% off billboards or full price billboards, it is considered that the type does not match.

[0073] In the process of intelligent inspection of store materials, by introducing standardized detection methods for store materials, it is possible to automatically inspect issues such as store material layout, avoid practical problems such as time-consuming and labor-intensive manual inspections and insufficiently unified inspection standards, realize intelligent inspections of retail stores, and improve enterprise operational efficiency.

[0074] Based on this, the present disclosure provides a standardized detection device for store materials, referring to Figure 5 As shown, Figure 5 It is a structural schematic diagram of a standardized detection device for store materials provided according to an embodiment of the present disclosure.

[0075] like Figure 5 As shown, the store material standard detection device 100 includes: a first processing unit 110, a second processing unit 120 and a judgment unit 130.

[0076] The first processing unit 110 is used to determine and crop at least one first material image from at least one store scene image of the store to be detected.

[0077] In some embodiments, determining and cropping at least one first material image from at least one store scene image of a store to be inspected includes: acquiring at least one store scene image; determining the coordinate information of each material in at least one store scene image based on a preset material detection model; and cropping at least one first material image from at least one store scene image based on the coordinate information of each material, wherein the first material image is a single material image with marked coordinate information.

[0078] The second processing unit 120 is used to compare the feature data of each cropped first material image with the feature data in a preset feature library based on a preset material classification model, and determine the material category of each first material image according to the feature comparison result.

[0079] In some embodiments, the feature data in the preset feature library includes feature data of each second material image in at least two angular directions, wherein the second material image is a single material image with a marked material category.

[0080] In some embodiments, the feature data in the preset feature library includes feature data of each second material image in directions of 0°, 90°, 180°, and 270°.

[0081] In some embodiments, the feature data of each cropped first material image is compared with the feature data in a preset feature library, including: extracting a multidimensional feature vector of the cropped first material image; based on the multidimensional feature vector of the first material image, determining the cosine similarity between the multidimensional feature vector of the first material image and the multidimensional feature vector of the compared second material image in at least two angular directions; and using the cosine similarity as the feature comparison result.

[0082] In some embodiments, the material category of each first material image is determined based on the feature comparison result, including: when at least one cosine similarity in the feature comparison result is greater than a first threshold, determining that the material category of the first material image is the material category of the second material image being compared; when all cosine similarities in the feature comparison result are not greater than the first threshold, comparing the feature data of the first material image with the feature data of the next second material image in at least two angular directions, and determining the material category of the first material image again based on the feature comparison result.

[0083] The judgment unit 130 is used to judge whether the material category of the store meets the specification based on the material category of each first material image and according to the preset material specification standard.

[0084] In some embodiments, based on the material category of each first material image and according to preset material specification standards, determining whether the material category of the store meets the specifications includes: determining all material categories of the store and the material quantities corresponding to each material category based on the material category of each first material image; determining whether all material categories of the store and the material quantities corresponding to each material category meet the specifications according to preset material specification standards, wherein the preset material specification standards include at least one material category and the material quantity corresponding to each material category.

[0085] The standardized detection device for store materials provided in the embodiments of the present disclosure adopts the standardized detection method for store materials in the above embodiments, which can solve the technical problems in the background technology.

[0086] The beneficial effects of the standardized detection device for store materials provided by the present disclosure are the same as the beneficial effects of the standardized detection method for store materials provided by the above-mentioned embodiments, and other technical features in the standardized detection device for store materials are the same as the disclosed features of the standardized detection method for store materials, which will not be repeated here.

[0087] Based on this, an embodiment of the present disclosure provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the method for standardized detection of store materials provided in the above embodiment.

[0088] The beneficial effects of the machine-readable storage medium provided in the present disclosure are the same as the beneficial effects of the standardized detection method for store materials provided in the above-mentioned embodiment, and will not be elaborated here.

[0089] An embodiment of the present disclosure also provides an electronic device, which includes: a memory; and a processor, wherein the processor is configured to execute the standardized detection method for store materials provided in the above embodiment.

[0090] In a typical configuration, a device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0091] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0092] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0093] The beneficial effects of the electronic device provided by the embodiment of the present disclosure are the same as the beneficial effects of the standardized detection method for store materials provided by the above-mentioned embodiment, and other technical features in the device are the same as the features disclosed in the above-mentioned embodiment method, which will not be repeated here.

[0094] The embodiments of the present disclosure also provide a computer program product, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for standardizing the detection of store materials.

[0095] The beneficial effects of the computer program product provided by the embodiments of the present disclosure are the same as the beneficial effects of the standardized detection method for store materials provided by the above embodiments, and will not be elaborated here.

[0096] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0097] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0098] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0099] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0100] It should be noted that although the expressions "first", "second", etc. are used herein to describe different modules, steps, data, etc. of the embodiments of the present disclosure, the expressions "first", "second", etc. are only used to distinguish between different modules, steps, data, etc., and do not indicate a specific order or importance. In fact, the expressions "first", "second", etc. can be used interchangeably.

[0101] Although operations are described in a particular order in the drawings, this should not be understood as requiring that the operations be performed in the particular order shown or in serial order, or that all shown operations be performed to achieve desired results. In certain circumstances, multitasking and parallel processing may be advantageous.

[0102] The acquisition, transmission, storage, use, and processing of data in the embodiments of the present disclosure are in compliance with the relevant provisions of national laws and regulations.

[0103] It should be noted that in the embodiments of the present disclosure, certain software, components, models and other existing solutions in the industry may be mentioned, which should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present disclosure, but it does not mean that the applicant has or will necessarily use the solution.

[0104] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0105] The above are only embodiments of the present disclosure and are not intended to limit the present disclosure. For those skilled in the art, the present disclosure may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present disclosure shall be included in the scope of the claims of the present disclosure.

Claims

1. A standardized testing method for store materials, characterized in that: The standard detection method includes: Determine and crop at least one first material image from at least one store scene image of the store to be inspected; Based on a preset material classification model, feature data of each cropped first material image is compared with feature data in a preset feature library, and a material category of each first material image is determined according to the feature comparison result; Based on the material category of each first material image and according to preset material specification standards, it is determined whether the material category of the store meets the specifications.

2. The standard detection method according to claim 1, characterized in that: The step of determining and cropping at least one first material image from at least one store scene image of the store to be detected includes: Acquire the at least one store scene image; Based on a preset material detection model, determining coordinate information of each material in the at least one store scene image; Based on the coordinate information of each material, at least one first material image is cropped from the at least one store scene image, wherein the first material image is a single material image with marked coordinate information.

3. The standard detection method according to claim 1 or 2, characterized in that: The feature data in the preset feature library includes feature data of each second material image in at least two angular directions, wherein the second material image is a single material image with a marked material category.

4. The standard detection method according to claim 3, characterized in that: The feature data in the preset feature library includes feature data of each second material image in directions of 0°, 90°, 180° and 270°.

5. The standard detection method according to claim 3, characterized in that: The step of comparing the feature data of each cropped first material image with the feature data in a preset feature library includes: Extracting a multidimensional feature vector of the cropped first material image; Based on the multidimensional feature vector of the first material image, determining the cosine similarity between the multidimensional feature vector of the first material image and the multidimensional feature vector of the second material image being compared in at least two angular directions; The cosine similarity is used as the feature comparison result.

6. The standard detection method according to claim 5, characterized in that: The determining the material category of each first material image according to the feature comparison result includes: When at least one cosine similarity in the feature comparison result is greater than a first threshold, determining the material category of the first material image as the material category of the compared second material image; When all cosine similarities in the feature comparison result are not greater than a first threshold, the feature data of the first material image is compared with the feature data of the next second material image in at least two angular directions, and the material category of the first material image is determined again based on the feature comparison result.

7. The standard detection method according to claim 1, characterized in that: Based on the material category of each first material image, judging whether the material category of the store meets the specification according to a preset material specification standard includes: Based on the material category of each first material image, determine all material categories of the store and the material quantity corresponding to each material category; According to the preset material specification standard, it is determined whether all material categories of the store and the material quantity corresponding to each material category meet the specification, and the preset material specification standard includes at least one material category and the material quantity corresponding to each material category.

8. A standardized testing device for store materials, characterized in that: The standard detection device comprises: A first processing unit, configured to determine and crop at least one first material image from at least one store scene image of the store to be detected; a second processing unit, configured to compare feature data of each cropped first material image with feature data in a preset feature library based on a preset material classification model, and determine a material category of each first material image according to a feature comparison result; A judgment unit is used to judge whether the material category of the store meets the specification based on the material category of each first material image and according to a preset material specification standard.

9. An electronic device, characterized in that: The electronic device comprises: Memory; and A processor, the processor being configured to execute the standardized detection method for store materials according to any one of claims 1-7.

10. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions for enabling the machine to execute the standardized detection method for store materials described in any one of claims 1-7.

Images