A method, apparatus and device for processing display cabinet images

Abstract

The application discloses a display cabinet image processing method, device and equipment, relates to the technical field of deep learning, and solves the technical problem that the display cabinet image exists in inclined shooting, which leads to low recognition accuracy of the display cabinet image. The method comprises the following steps: S100, a display cabinet image is acquired, the cabinet grid of the display cabinet image is detected, and a plurality of detection boxes are obtained; S200, according to the plurality of detection boxes, the plurality of cabinet grids of the display cabinet image are grouped longitudinally, and a plurality of longitudinal groups are obtained; S300, according to the plurality of longitudinal groups, the plurality of cabinet grids of the display cabinet image are grouped transversely, and a plurality of transverse groups are obtained; and S400, according to the plurality of longitudinal groups and the plurality of transverse groups, a cabinet grid display image of the display cabinet image is obtained. The application is used for providing a display cabinet image processing method, device and equipment, correcting the display cabinet image in inclined shooting, and obtaining a cabinet grid display image with regular display.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of deep learning, and in particular to a display cabinet image processing method, device and equipment. BACKGROUND

[0002] With the rapid development of science and technology, target detection has been paid more and more attention, and target detection has been integrated into various aspects of life. For example, the retail industry only needs a picture to classify the product specifications of the goods appearing in the picture, avoiding the cumbersome process of manual review.

[0003] Now, through target detection technology, the display cabinet image is recognized, so as to identify the placement position or the number of goods in the cabinet, so that the administrator can efficiently manage the cabinet and the goods. For example, the cabinet lacks a certain good, and the target detection technology identifies the missing goods, so that the administrator can replenish the goods in time. However, for a large number of display cabinet images, due to the internal structure of different cabinets and the shooting angle of the display cabinet image and other problems, the recognition accuracy of the display cabinet image is low, resulting in inefficient management of the cabinet and the goods.

[0004] In the process of implementing the present application, the inventors have found that the prior art at least has the following problems:

[0005] Due to the tilted shooting of the display cabinet image, the recognition accuracy of the display cabinet image is low. SUMMARY

[0006] The present application aims to provide a display cabinet image processing method, device and equipment to solve the technical problem that the display cabinet image in the prior art has a tilted shooting, resulting in low recognition accuracy of the display cabinet image. The preferred technical solutions in the many technical solutions provided by the present application can produce many technical effects, which are described in detail below.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0008] The present application provides a display cabinet image processing method, comprising:

[0009] S100: acquiring a display cabinet image, detecting the cabinet compartments of the display cabinet image, and obtaining a plurality of detection boxes;

[0010] S200: grouping the plurality of cabinet compartments of the display cabinet image in a vertical direction according to the plurality of detection boxes, and obtaining a plurality of vertical groups;

[0011] S300: grouping the plurality of cabinet compartments of the display cabinet image in a horizontal direction according to the plurality of vertical groups, and obtaining a plurality of horizontal groups;

[0012] S400: obtaining a shelf arrangement diagram of the display cabinet image according to the plurality of longitudinal groups and the plurality of transverse groups.

[0013] Preferably, in the S200 step, the longitudinal grouping of the plurality of shelves of the display cabinet image specifically comprises:

[0014] S210: obtaining a horizontal coordinate projection of each of the bounding boxes;

[0015] S220: obtaining a longitudinal intersection-over-union value of any two bounding boxes according to the horizontal coordinate projections of the two bounding boxes;

[0016] S230: if the longitudinal intersection-over-union value of the two bounding boxes is greater than a longitudinal grouping threshold, determining the two bounding boxes as a same longitudinal group.

[0017] Preferably, in the S300 step, the transverse grouping of the plurality of shelves of the display cabinet image according to the plurality of longitudinal groups comprises:

[0018] sequentially arranging the plurality of longitudinal groups according to a preset sequence, matching an i-th bounding box of any longitudinal group with a j-th bounding box of any adjacent group of the longitudinal group to obtain a matching result of the i-th bounding box and the j-th bounding box, wherein i and j are positive integers greater than 0;

[0019] transversely grouping the plurality of bounding boxes according to the matching result of the i-th bounding box and the j-th bounding box to obtain the plurality of transverse groups.

[0020] Preferably, the matching of the i-th bounding box of any longitudinal group with the j-th bounding box of any adjacent group of the longitudinal group comprises:

[0021] S310: obtaining a vertical coordinate projection of each of the bounding boxes in each of the longitudinal groups;

[0022] S320: obtaining a transverse intersection-over-union value of any two bounding boxes of any two adjacent longitudinal groups according to the vertical coordinate projections of the two longitudinal groups;

[0023] S330: matching the transverse intersection-over-union value of any two bounding boxes of any two adjacent longitudinal groups according to a Hungarian algorithm to determine whether the two bounding boxes of the two adjacent longitudinal groups are a same transverse group.

[0024] Preferably, the transverse intersection-over-union value is calculated according to the following formula:

[0025] G k={B1(y1,y2),B2(y1,y2),....,Bn(y1,y2)};

[0026]

[0027] wherein, IOUy is the horizontal intersection over union value; Gk represents a set of the vertical coordinate projections of each of the bounding boxes in the kth group; Bn(y1, y2) is the vertical coordinate projection of the nth bounding box, and n and k are positive integers greater than 0.

[0028] Preferably, the calculation formula of the vertical intersection over union value is as follows:

[0029]

[0030] wherein, IOUx is the vertical intersection over union value; Bn(x1, x2) and Bm(x1, x2) represent the horizontal coordinate projections of the nth and mth bounding boxes respectively, and n and m are positive integers greater than 0.

[0031] Preferably, if the vertical intersection over union value of any two bounding boxes is less than the vertical grouping threshold, the two bounding boxes are determined as different vertical groups.

[0032] Preferably, in the S400 step, the multiple vertical groups and the multiple horizontal groups are sorted in a preset order in a two-dimensional coordinate system to obtain a cabinet grid arrangement diagram of the display cabinet image.

[0033] The application further provides a display cabinet image processing device, which comprises:

[0034] an input module configured to input a display cabinet image and extract multiple bounding boxes;

[0035] a vertical grouping module configured to group the multiple bounding boxes vertically to obtain multiple vertical groups;

[0036] a horizontal grouping module configured to group the multiple vertical groups horizontally to obtain multiple horizontal groups;

[0037] an output module configured to sort the vertical groups and the horizontal groups and output a cabinet grid arrangement diagram of the display cabinet image.

[0038] In addition, the application further provides a display cabinet image processing device, which comprises:

[0039] one or more processors;

[0040] The memory is configured to store one or more computer programs, and the one or more processors are configured to execute the one or more computer programs stored in the memory, so that the one or more processors execute the display cabinet image processing method.

[0041] The above technical solutions of the present application have the following advantages or beneficial effects:

[0042] The present application extracts multiple cabinet detection frames of the display cabinet image, and first groups the detection frames in the longitudinal direction by using the intersection-over-union threshold value comparison to obtain multiple longitudinal groups. Then, the multiple detection frames in any two adjacent longitudinal groups are grouped in the horizontal direction by using the Hungarian algorithm to obtain multiple horizontal groups. In combination with the longitudinal groups and the horizontal groups, the display cabinet image that is tilted and shot can be corrected to obtain a regular cabinet display image. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained according to these drawings without creative labor, and the drawings are as follows:

[0044] Figure 1 is a method flowchart of the first embodiment of the present application;

[0045] Figure 2 is a step flowchart of S200 of the first embodiment of the present application;

[0046] Figure 3 is a step flowchart of S300 of the first embodiment of the present application;

[0047] Figure 4 is a structural schematic diagram of the second embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the present application more clear, the various exemplary embodiments to be described below will be described with reference to the corresponding drawings, which constitute a part of the exemplary embodiments, and various exemplary embodiments that can be used to implement the present application are described. Unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation described in the following exemplary embodiments does not represent all the implementations consistent with the present disclosure. It should be understood that they are only examples of processes, methods and devices, etc. consistent with some aspects of the present disclosure as described in the appended claims, and other embodiments can also be used, or modifications can be made to the embodiments listed herein in structure and function without departing from the scope and spirit of the present application.

[0049] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse" and the like indicate the orientation or positional relationship shown in the drawings, only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the elements referred to must have a specific orientation, be constructed and operated in a specific orientation. The terms "first", "second" and the like are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. The term "multiple" means two or more. The terms "connected", "connected" should be broadly understood, for example, it can be fixed connection, detachable connection, integral connection, mechanical connection, electrical connection, communication connection, direct connection, indirect connection through intermediate medium, internal communication of two elements or interaction relationship between two elements. The term "and / or" includes any and all combinations of one or more related listed items. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0050] In order to illustrate the technical solutions described in the present application, the following will be described by specific examples, only showing the part related to the embodiment of the present application.

[0051] Embodiment one:

[0052] As Figure 1 shown, the present application provides a display cabinet image processing method, comprising: S100: acquiring a display cabinet image, detecting the cabinet of the display cabinet image, and obtaining a plurality of detection boxes; S200: grouping the plurality of cabinet of the display cabinet image according to the plurality of detection boxes, and obtaining a plurality of longitudinal groups; S300: grouping the plurality of cabinet of the display cabinet image according to the plurality of longitudinal groups, and obtaining a plurality of transverse groups; S400: obtaining the cabinet arrangement diagram of the display cabinet image according to the plurality of longitudinal groups and the plurality of transverse groups. Specifically, the present application uses yolov5s model to detect the target of the acquired display cabinet image, and can obtain the detection box of the plurality of cabinet. Then the left upper corner coordinate (x1, y1) and the right upper corner coordinate (x2, y2) of each cabinet detection box are extracted, so as to obtain the projection information of each detection box on x axis and y axis respectively.

[0053] Further, the present application extracts the plurality of cabinet detection boxes of the display cabinet image, first groups the detection boxes in the longitudinal direction by comparing the intersection over union threshold value, and obtains a plurality of longitudinal groups. Then, the plurality of detection boxes in any two adjacent longitudinal groups are grouped in the transverse direction by using the Hungarian algorithm, and a plurality of transverse groups are obtained. Combined with the longitudinal group and the transverse group, the display cabinet image tilted and shot can be corrected, and a regular display cabinet arrangement image is obtained.

[0054] AsFigure 2 As shown, as an optional embodiment, in the S200 step, the multiple compartments of the display cabinet image are vertically grouped, specifically comprising: S210: obtaining the horizontal coordinate projection of each detection box; S220: obtaining the vertical intersection-over-union value of any two detection boxes according to the horizontal coordinate projections of the two detection boxes; S230: if the vertical intersection-over-union value of any two detection boxes is greater than the vertical grouping threshold, the two detection boxes are determined as the same vertical group. Specifically, in the S200 step, since the top-left corner coordinate (x1, y1) and the top-right corner coordinate (x2, y2) of each compartment detection box have been extracted in the step S100, the horizontal coordinate projection of each detection box is the projection length represented by the x-axis (x1, x2) in the two-dimensional coordinate system. The horizontal coordinate projection of any two detection boxes can be calculated according to the calculation formula of the vertical intersection-over-union value. As long as the vertical intersection-over-union value of any two detection boxes is compared with the vertical grouping threshold, if the vertical intersection-over-union value of the two detection boxes is greater than the vertical grouping threshold, the two detection boxes are determined as the same vertical group.

[0055] Further, in order to improve the efficiency of vertically grouping the multiple detection boxes, a pre-sorting can be performed. First, the center point and the horizontal coordinate projection of each detection box need to be obtained, and the center point matches the horizontal coordinate projection. Then, according to the center point of each detection box, each detection box is sorted in a preset order to obtain a sorted detection box set. The preset order can be set by the user. Then, the detection box set is traversed, and the vertical intersection-over-union value of the detection boxes is calculated two by two through the horizontal coordinate projection. If the vertical intersection-over-union value is greater than the vertical grouping threshold, the two detection boxes are determined as the same vertical group. In this embodiment, the pre-sorting and the preset order can be selected as sorting and traversing from high to low and from left to right.

[0056] As an optional embodiment, the calculation formula of the vertical intersection-over-union value is:

[0057]

[0058] wherein, IOUx is the vertical intersection-over-union value; Bn(x1, x2) and Bm(x1, x2) represent the horizontal coordinate projection of the nth and mth detection boxes respectively, and n and m are positive integers greater than 0. Specifically, by projecting the top-left corner coordinate (x1, y1) and the top-right corner coordinate (x2, y2) of each compartment detection box on the x-axis, the horizontal coordinate projection (x1, x2) of each detection box can be obtained. Further, the intersection-over-union (IoU) is a concept used in target detection, which refers to the overlap rate between two detection boxes, i.e., the ratio of their intersection to union.

[0059] As an optional implementation, if the vertical intersection-over-union value of any two detection boxes is less than the vertical grouping threshold, the two detection boxes are determined as different vertical groups. Specifically, comparing the vertical intersection-over-union value of any two detection boxes with the vertical grouping threshold can confirm whether the two detection boxes belong to the same vertical group. If less than the threshold, the two detection boxes do not belong to the same vertical group.

[0060] As an optional implementation, in the step S300, the multiple cabinet compartments of the display cabinet image are horizontally grouped according to the multiple vertical groups as follows: the multiple vertical groups are sorted according to a preset order, the i-th detection box of any vertical group is matched with the j-th detection box of any adjacent group to obtain a matching result of the i-th detection box and the j-th detection box, where i and j are positive integers greater than 0; and the multiple detection boxes are horizontally grouped according to the matching result of the i-th detection box and the j-th detection box to obtain multiple horizontal groups. Specifically, in order to horizontally group the multiple vertical groups, the horizontal intersection-over-union value of any two detection boxes of any two adjacent vertical groups needs to be obtained. Then, according to the horizontal intersection-over-union value of the two detection boxes, the Hungarian algorithm is used to one-to-one match any two detection boxes in the two adjacent vertical groups, and the two detection boxes that are successfully matched are classified into the same horizontal group. The Hungarian algorithm is a combinatorial optimization algorithm for solving task assignment problems in polynomial time, and promotes the later primal-dual method, which is mainly used to solve some problems related to bipartite graph matching. The Hungarian algorithm is applied in the present application to find the optimal matching of each detection box in the vertical group in the horizontal grouping.

[0061] As Figure 3As shown, as an optional embodiment, matching the ith detection frame of any longitudinal group with the jth detection frame of any adjacent group is matched as: S310: obtaining the longitudinal coordinate projection of each detection frame in each longitudinal group; S320: obtaining the horizontal intersection-over-union value of any two detection frames of any two adjacent longitudinal groups according to the longitudinal coordinate projections of the two groups; S320: matching the horizontal intersection-over-union value of any two detection frames of any two adjacent longitudinal groups according to the Hungarian algorithm to determine whether the two detection frames are in the same horizontal group. Specifically, in this embodiment, the plurality of longitudinal groups are first sorted according to the preset order from left to right to obtain the sorted plurality of longitudinal groups. Optionally, the first longitudinal group and the second longitudinal group are adjacent longitudinal groups in the sorted longitudinal groups. Then, the horizontal intersection-over-union value of any two detection frames in the first longitudinal group and the second longitudinal group is obtained, and an intersection-over-union matrix M is established according to the first longitudinal group and the second longitudinal group. The matrix value is the horizontal intersection-over-union value of any two detection frames in the first longitudinal group and the second longitudinal group. The intersection-over-union matrix M is input into the Hungarian algorithm, and the optimal matching result of any two detection frames in the first longitudinal group and the second longitudinal group can be obtained by matching through the algorithm. The adjacent groups of the sorted longitudinal groups are matched in a loop, and whether any two detection frames in the plurality of adjacent groups are in the same horizontal group can be determined.

[0062] According to the matching result of the Hungarian algorithm, the detection frames that are successfully matched in the two adjacent longitudinal groups are classified into the same horizontal group. Then, the detection frames that are not successfully matched are selected and added to the adjacent group of the longitudinal group. When the adjacent group and the next longitudinal group adjacent thereto are matched, the horizontal grouping is performed again.

[0063] It can be explained that the detection frames in the first longitudinal group that are not successfully matched with the second longitudinal group are integrated into the second longitudinal group. When the Hungarian algorithm is used to match the second longitudinal group and the third longitudinal group, the horizontal grouping is performed again. Finally, the one-to-one adjacent group matching process is cycled until all the detection frames of all the longitudinal groups are matched.

[0064] It should be noted that the above-mentioned adjacent matching of the first longitudinal group and the second longitudinal group is only the most preferred option of this embodiment. After obtaining the plurality of longitudinal groups, the Hungarian algorithm can be used to match any two adjacent longitudinal groups, and then the adjacent matching between other adjacent longitudinal groups can be performed. According to the one-to-one adjacent group matching process, even if an optional longitudinal group is selected as the starting point and matched with the adjacent longitudinal group, the horizontal grouping of all the detection frames of all the longitudinal groups can also be completed.

[0065] As an optional embodiment, the calculation formula of the horizontal intersection-over-union value is:

[0066] Gk ={B1(y1,y2),B2(y1,y2),....,Bn(y1,y2)};

[0067]

[0068] wherein, IOUy is the horizontal intersection over union value; Gk represents the set of vertical coordinate projections of each detection box in the kth group; Bn(y1, y2) is the vertical coordinate projection of the nth detection box, and n and k are positive integers greater than 0. Specifically, by projecting the left upper corner coordinate (x1, y1) and the right upper corner coordinate (x2, y2) of each cabinet compartment detection box on the y-axis, the vertical coordinate projection (y1, y2) of each detection box can be obtained.

[0069] It should be noted that when calculating in this step, the vertical coordinate projection of each vertical group detection box is set up as a set. When performing the Hungarian algorithm matching, the intersection over union matrix M of the vertical coordinate projection set of the vertical group and the vertical coordinate projection set of the adjacent vertical group is calculated, which is used as the input of the Hungarian algorithm, and the optimal matching result is obtained after the algorithm matching.

[0070] As an optional embodiment, in the step S400, the cabinet compartment arrangement graph of the display cabinet image is obtained by sorting the multiple vertical groups and the multiple horizontal groups in a preset order in the two-dimensional coordinate system. Specifically, the detection boxes are sorted according to the vertical groups and the horizontal groups in the two-dimensional coordinate system, and after the sorting is completed, the cabinet compartment arrangement graph of the display cabinet image is saved.

[0071] The embodiment is only a specific example, and does not indicate that the present application is only in this implementation manner.

[0072] Embodiment two:

[0073] As shown in Figure 4 , the present application further provides a display cabinet image processing device, which comprises: an input module for inputting a display cabinet image and extracting multiple detection boxes; a vertical grouping module for vertically grouping the multiple detection boxes to obtain multiple vertical groups; a horizontal grouping module for horizontally grouping the multiple vertical groups to obtain multiple horizontal groups; and an output module for sorting the vertical groups and the horizontal groups and outputting a cabinet compartment arrangement graph of the display cabinet image. Specifically, the step of outputting the display cabinet image taken at an angle into a cabinet compartment arrangement graph of a regular arrangement is as described in the embodiment one.

[0074] Embodiment three:

[0075] The application further provides an embodiment of a display cabinet image processing device, comprising one or more processors and a memory; wherein the memory is configured to store one or more computer programs, and the one or more processors are configured to execute the one or more computer programs stored in the memory, so that the processor executes the features / steps of the above-mentioned embodiment one.

[0076] The above is only the preferred embodiment of the present application, and those skilled in the art know that various changes or equivalent replacements can be made to the features and embodiments without departing from the spirit and scope of the present application. In addition, the features and embodiments can be modified to adapt to specific conditions and materials under the guidance of the present application without departing from the spirit and scope of the present application. Therefore, the present application is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of the present application are within the protection scope of the present application.

Claims

1. A method for processing images of display cabinets, characterized in that, include: S100: Acquire an image of the display cabinet, detect the cabinet cells in the image of the display cabinet, and obtain multiple detection boxes; S200: Based on the multiple detection frames, the multiple cabinet compartments of the display cabinet image are vertically grouped to obtain multiple vertical groups; S300: Based on the plurality of vertical groups, the plurality of cabinet compartments of the display cabinet image are horizontally grouped to obtain a plurality of horizontal groups; S400: Based on the plurality of vertical groups and the plurality of horizontal groups, obtain the display cabinet layout diagram of the display cabinet image; In step S200, the vertical grouping of the multiple display cabinets in the display cabinet image specifically includes: S210: Obtain the horizontal coordinate projection of each detection box; S220: Based on the horizontal coordinate projection of any two detection boxes, obtain the vertical intersection-union ratio of the two detection boxes; S230: If the vertical intersection-union ratio of any two detection boxes is greater than the vertical grouping threshold, then the two detection boxes are determined to be in the same vertical group; In step S300, the step of horizontally grouping the multiple cabinet compartments of the display cabinet image according to the multiple vertical groups is as follows: The plurality of vertical groups are sorted according to a preset order, and the i-th detection box of any vertical group is matched with the j-th detection box of any adjacent group to obtain the matching result of the i-th detection box and the j-th detection box, where i and j are positive integers greater than 0; Based on the matching result between the i-th detection box and the j-th detection box, the multiple detection boxes are horizontally grouped to obtain multiple horizontal groups.

2. The method for processing display cabinet images according to claim 1, characterized in that, The matching process involves matching the i-th detection box of any vertical group with the j-th detection box of any adjacent group as follows: S310: Obtain the ordinate projection of each detection box in each of the vertical groups; S320: Based on the ordinate projection of any two adjacent vertical groups, obtain the lateral intersection-union ratio of any two detection boxes in any two adjacent vertical groups; S330: Match the lateral intersection-union ratio of any two detection boxes in any two adjacent vertical groups according to the Hungarian algorithm to determine whether any two detection boxes in any two adjacent vertical groups belong to the same horizontal group.

3. The method for processing display cabinet images according to claim 2, characterized in that, The formula for calculating the lateral intersection ratio is: ； ； Where IOUy is the horizontal intersection-union ratio; Gk represents the set of the ordinate projections of each detection box in the kth group; Bn(y1, y2) is the ordinate projection of the nth detection box, Bm(y1, y2) is the ordinate projection of the mth detection box, and n, k, and m are positive integers greater than 0.

4. The method for processing display cabinet images according to claim 1, characterized in that, The formula for calculating the longitudinal intersection-union ratio is: ； Among them, IOU x The vertical intersection-union ratio is given; Bn(x1, x2) and Bm(x1, x2) represent the horizontal coordinate projections of the nth and mth detection boxes, respectively, where n and m are positive integers greater than 0.

5. The method for processing display cabinet images according to claim 1, characterized in that, If the vertical intersection-union ratio of any two detection boxes is less than the vertical grouping threshold, then the two detection boxes are determined to be different vertical groups.

6. The method for processing display cabinet images according to claim 1, characterized in that, In step S400, the multiple vertical groups and the multiple horizontal groups are sorted in a preset order in a two-dimensional coordinate system to obtain the display cabinet image of the display cabinet.

7. A device for processing images of display cabinets, characterized in that, include: The input module is used to input images of the display cabinets and extract multiple detection boxes; The vertical grouping module is used to vertically group the multiple detection boxes to obtain multiple vertical groups. The vertical grouping module obtains the horizontal coordinate projection of each detection box, and obtains the vertical intersection-union ratio of any two detection boxes based on the horizontal coordinate projection of any two detection boxes. If the vertical intersection-union ratio of any two detection boxes is greater than the vertical grouping threshold, then the two detection boxes are determined to be in the same vertical group. A horizontal grouping module is used to horizontally group the multiple vertical groups to obtain multiple horizontal groups. The horizontal grouping module sorts the multiple vertical groups according to a preset order, and matches the i-th detection box of any vertical group with the j-th detection box of any adjacent group to obtain the matching result of the i-th detection box and the j-th detection box, where i and j are positive integers greater than 0. Based on the matching result of the i-th detection box and the j-th detection box, the multiple detection boxes are horizontally grouped to obtain multiple horizontal groups. The output module is used to sort the vertical group and the horizontal group, and output the display cabinet image as a cabinet display diagram.

8. A device for processing images of display cabinets, characterized in that, include: One or more processors; A memory for storing one or more computer programs, and one or more processors for executing the one or more computer programs stored in the memory to cause the one or more processors to perform a method for processing display cabinet images as described in any one of claims 1-6.

Images