Store area determination method, apparatus and device, readable storage medium and product

Abstract

The invention provides a store area determination method, device and equipment, a readable storage medium and a product, and relates to the field of image processing, in particular to the field of deep learning. According to the specific implementation scheme, the method comprises the steps of obtaining a to-be-tested image associated with a target store, and determining a content category corresponding to the to-be-tested image through a preset first classification model; determining a target object associated with the to-be-detected image by adopting a target detection model matched with the content category; a target mapping list matched with the target object is determined, the target mapping list comprises a plurality of mapping relation pairs, and each mapping relation pair comprises the target object and an area interval corresponding to the target object; and determining a target area interval corresponding to the target store based on the target mapping list and the target object. Therefore, the coverage area interval of the target store can be quickly and accurately estimated based on the to-be-detected image associated with the target store, and the efficiency and accuracy of store area determination are improved.

Description

Technical Field

[0001] The present disclosure relates to deep learning in image processing, and more particularly to a method, device, equipment, readable storage medium and product for determining store area. Background Art

[0002] Fast-moving consumer goods (FMCG) refer to consumer goods with a short lifespan and a fast consumption rate. FMCG brands generally place their products in multiple stores so that users can easily obtain them. Therefore, the determination of store coverage is often crucial for the placement of FMCG products and the content display of stores selling FMCG products in online maps.

[0003] Therefore, how to quickly and accurately determine the coverage area of ​​stores selling fast-moving consumer goods has become an urgent problem to be solved. Summary of the invention

[0004] The present disclosure provides a store area determination method, device, equipment, readable storage medium and product for determining store coverage area intervals based on store associated images.

[0005] According to a first aspect of the present disclosure, a method for determining a store area is provided, comprising:

[0006] Acquire the image to be tested associated with the target store, and determine the content category corresponding to the image to be tested by using a preset first classification model;

[0007] Determine the target object associated with the image to be detected by using a target detection model that matches the content category;

[0008] Determine a target mapping list matching the target object, wherein the target mapping list includes a plurality of mapping relationship pairs, and the mapping relationship pairs include the target object and an area interval corresponding to the target object;

[0009] A target area interval corresponding to the target store is determined based on the target mapping list and the target object.

[0010] According to a second aspect of the present disclosure, a store area determination device is provided, comprising:

[0011] An acquisition module, used to acquire an image to be tested associated with a target store, and determine a content category corresponding to the image to be tested by using a preset first classification model;

[0012] A determination module, configured to determine a target object associated with the image to be detected by using a target detection model that matches the content category;

[0013] A mapping module, used to determine a target mapping list matching the target object, wherein the target mapping list includes a plurality of mapping relationship pairs, and the mapping relationship pairs include the target object and an area interval corresponding to the target object;

[0014] A processing module is used to determine a target area interval corresponding to the target store based on the target mapping list and the target object.

[0015] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect.

[0019] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to execute the method described in the first aspect.

[0020] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising: a computer program, wherein the computer program is stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program so that the electronic device executes the method described in the first aspect.

[0021] The technology disclosed herein improves the efficiency and accuracy of store area determination.

[0022] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0024] Figure 1 A diagram of the system architecture on which the present disclosure is based;

[0025] Figure 2 A flow chart of a method for determining store area provided in an embodiment of the present disclosure;

[0026] Figure 3A flowchart of a method for determining store area provided in yet another embodiment of the present disclosure;

[0027] Figure 4 A schematic diagram of an application scenario provided for an embodiment of the present disclosure;

[0028] Figure 5 A schematic diagram of a flow chart of a method for determining store area provided in yet another embodiment of the present disclosure;

[0029] Figure 6 A schematic diagram of the structure of a store area determination device provided in an embodiment of the present disclosure;

[0030] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0031] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0032] The present disclosure provides a store area determination method, device, equipment, readable storage medium and product, which are applied to deep learning in the field of image processing to achieve the effect of predicting the coverage area interval of a store based on store images.

[0033] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0034] Fast-moving consumer goods (FMCG) refer to consumer goods with a short lifespan and a fast consumption rate. FMCG brands generally place their products in multiple stores so that users can easily obtain them. In order to facilitate the subsequent placement of goods, each FMCG brand will have its own store database, which records which stores the brand has sold its products to and on which store shelves. In addition, online maps often record parameter information of multiple stores, such as store locations, store coverage areas, etc. Users can achieve a preliminary understanding of store information through online maps.

[0035] Therefore, determining the store’s coverage area is often crucial to the launch of fast-moving consumer goods and the display of store content in online maps.

[0036] In the related art, stores are generally surveyed manually on the spot to determine the store coverage area. However, the above method for determining the store area is often inefficient and inaccurate.

[0037] In the process of solving the above technical problems, the inventors discovered through research that in order to quickly and accurately determine the coverage area range of a store, image processing can be performed by combining image classification and target detection to determine the area range of the store.

[0038] Optionally, an image to be tested of a store can be obtained, and the image to be tested can be a real scene image of the store, or it can be a logo image of a chain store, etc. After the image to be tested is obtained, the target object associated with the image to be tested can be determined by combining image classification and target detection, wherein the target object includes but is not limited to the logo image of the chain store, the door opening of the store, and the classification information corresponding to the store, etc.

[0039] In addition, a mapping relationship between a target object and an area interval may be established in advance. Therefore, after determining the target object associated with a store, a mapping list corresponding to the target object may be determined. The area interval of the store may be determined based on the mapping list.

[0040] In order to enable readers to more deeply understand the implementation principle of the present disclosure, the following Figure 1-Figure 7 The embodiments of the present disclosure are further refined.

[0041] Figure 1 This is the system architecture diagram on which the present disclosure is based, such as Figure 1 As shown, the system architecture based on the present disclosure at least includes a terminal device 11, a server 12 and a data server 13. Among them, a store area determination device is set in the server 12. The store area determination device can be written in languages ​​such as C / C++, Java, Shell or Python; the terminal device 11 can be, for example, a mobile phone, a tablet computer, etc. The data server 13 can be a cloud server or a server cluster, which stores a large amount of data.

[0042] The server 12 can be connected to the terminal device 11 and the data server 13 in communication. Therefore, the image to be tested associated with the target store can be obtained from the terminal device 11. The target area interval of the target store is determined by combining image classification and target detection. The identification information of the target store is associated with the target area interval and stored in the data server 13 for subsequent data call.

[0043] Figure 2 A flow chart of a method for determining store area provided in an embodiment of the present disclosure is shown in FIG. Figure 2 As shown, the method includes:

[0044] Step 201: Acquire an image to be tested associated with a target store, and determine a content category corresponding to the image to be tested by using a preset first classification model.

[0045] The execution subject of this embodiment is a store area determination device, which can be coupled to a server.

[0046] In this embodiment, in order to quickly and accurately determine the target area interval of the target store, the image to be tested associated with the target store can be obtained. The image to be tested can be collected and uploaded by the user in real time at the target store, or the image to be tested can be pre-stored in the store data set associated with the target store.

[0047] The image to be tested may be an image taken outside the target store, which may include the target store's door, window, store logo, and other contents. Alternatively, the image to be tested may also be an image taken inside the target store, which may include the target store's in-store displays, shelves, store logos, and merchandise. Alternatively, the image to be tested may also be a store logo image associated with the target store. For example, if the target store is a store under a chain brand, the image to be tested may be the identification information corresponding to the chain brand, such as the name, logo, and other contents.

[0048] Furthermore, after the image to be tested is acquired, a preliminary classification operation may be performed on the image to be tested using a preset first classification model.

[0049] Optionally, the image to be tested may be input into the first classification model, and the content category associated with the image to be tested may be determined based on the output content of the first classification model, wherein the content category includes but is not limited to a store logo category, a store image category, and an irrelevant image category.

[0050] It should be noted that, in order to improve the accuracy of subsequent data processing, if it is determined based on the first classification model that the image to be tested is an image that is not related to the store, the image to be tested may be discarded.

[0051] Step 202: Use a target detection model that matches the content category to determine a target object associated with the image to be detected.

[0052] In this embodiment, the image to be tested may be associated with a target object, wherein the target object includes but is not limited to store identification information associated with the target store, door opening quantity parameters corresponding to the target store, and classification information corresponding to the target store.

[0053] Optionally, the store identification information may be an identification of the brand to which the target store belongs, such as a name, logo, etc. The door opening includes but is not limited to a display window, an entrance door, etc. corresponding to the target store. The classification information includes but is not limited to large supermarket categories, supermarket categories in shopping malls, convenience store categories, etc.

[0054] Furthermore, the classification model can be combined with the target detection model to accurately realize the target object corresponding to the image to be detected. For different content categories, different target detection models can be used to realize the detection operation of the target object.

[0055] For example, if the image to be tested is determined to be a store identification category, and its corresponding content is relatively simple, the store identification information corresponding to the image to be tested can be identified through the first target detection model. If the image to be tested is determined to be a store image category, since the image to be tested contains relatively rich content, its corresponding target object can be store identification information or door opening information or classification information. Therefore, the detection operation of the target object can be achieved by combining the first target detection model, the second target detection model and the second classification model.

[0056] Step 203: Determine a target mapping list that matches the target object, wherein the target mapping list includes a plurality of mapping relationship pairs, and the mapping relationship pairs include the target object and the area interval corresponding to the target object.

[0057] In this embodiment, in order to achieve accurate estimation of the area range of the target store, a mapping list corresponding to multiple different target objects may be pre-constructed.

[0058] Optionally, since the target object includes store identification information, door opening quantity parameters and classification information respectively, the mapping list may include a first mapping list, a second mapping list and a third mapping list. The first mapping list includes mapping relationships between different store identification information and area intervals. The second mapping list includes mapping relationships between different door opening quantity parameters and area intervals. The third mapping list includes mapping relationships between different classification information and area intervals.

[0059] Therefore, after determining the target object associated with the image to be tested, a target mapping list matching the target object can be determined. The target mapping list includes multiple mapping relationship pairs, and the mapping relationship pairs include the target object and the area interval corresponding to the target object.

[0060] Step 204: Determine a target area interval corresponding to the target store based on the target mapping list and the target object.

[0061] In this embodiment, after the target object and the target mapping list are determined respectively, the area interval matching the target object can be determined in the target mapping list and determined as the target area interval corresponding to the target store.

[0062] The store area determination method provided in this embodiment, after acquiring the image to be tested associated with the target store, identifies the target object in the image to be tested based on the combination of the classification model and the target detection model, and pre-sets the target mapping list associated with the target object, so that the target area interval corresponding to the target store can be quickly and accurately determined based on the target object and the target mapping list. Therefore, the target area interval of the target store can be determined based on an image to be tested associated with the target store, which improves the efficiency of determining the target area interval, does not require the user to conduct on-site surveys, and saves human resources.

[0063] Further, based on any of the above embodiments, after step 204, the method further includes:

[0064] The identification information of the target store and the target area interval are associated and stored in a preset storage path.

[0065] or,

[0066] Based on the target area interval associated with the target store, data updating operation is performed on the associated parameters of the target store in the map software.

[0067] In this embodiment, after the target area interval of the target store is determined based on the image to be tested, the identification information of the target store and the target area interval can be associated and stored in a preset storage path. For example, the identification information of the target store and the target area interval can be associated and stored in a preset store database. Therefore, the subsequent fast-moving consumer goods brand can determine the target area interval of the store based on the store database, so as to more flexibly and accurately place fast-moving consumer goods.

[0068] As an implementable manner, the associated parameters of the target store in the map software may be updated based on the target area interval associated with the target store. For example, if the area of ​​the target store displayed in the online map is different from the target area interval, the associated parameters of the target store in the map software may be updated based on the target area interval.

[0069] The store area determination method provided in this embodiment associates and stores the identification information of the target store and the target area interval, so that the subsequent fast-moving consumer goods brand can better place fast-moving consumer goods based on the parameter information of the target store. By updating the associated parameters of the target store in the map software based on the target area interval associated with the target store, subsequent users can obtain more accurate store information through the online map, thereby improving the user experience.

[0070] Optionally, based on any of the above embodiments, step 201 includes:

[0071] The image to be tested is input into a preset first classification model to obtain probability information output by the first classification model, where the probability information includes prediction probabilities corresponding to a plurality of preset categories.

[0072] The preset category whose predicted probability meets the preset condition is determined as the content category corresponding to the image to be tested.

[0073] In this embodiment, a first classification model may be preset, and the first classification model can accurately realize the content category corresponding to the image to be tested.

[0074] Optionally, after acquiring the image to be tested, the image to be tested can be input into a preset first classification model to obtain probability information output by the first classification model, wherein the probability information includes prediction probabilities corresponding to a plurality of preset categories. Therefore, the preset category whose prediction probability meets the preset conditions can be determined as the content category corresponding to the image to be tested.

[0075] For example, the preset category with the highest prediction probability can be determined as the content category corresponding to the image to be tested. The first classification model can output that the probability that the image to be tested belongs to the store logo category is 60%, the probability that the image to be tested belongs to the store image category is 30%, and the probability that it belongs to an irrelevant image type is 10%. Therefore, the content category of the image to be tested can be determined as the store logo category.

[0076] Optionally, if the content category of the image to be tested is determined to be an irrelevant image category based on the probability information output by the first classification model, the image to be tested may be discarded, and the user may be prompted to re-collect the image to be tested.

[0077] The store area determination method provided in this embodiment pre-trains the first classification model, so that after acquiring the image to be tested, the content category associated with the image to be tested can be accurately determined based on the first classification model, and then the corresponding target detection model can be used based on the content category to achieve accurate recognition of the target object.

[0078] Further, based on any of the above embodiments, before step 201, the method further includes:

[0079] A first training data set is obtained, wherein the first training data set includes a plurality of first training data pairs, wherein the first training data pairs include a first training image and annotation information associated with the first training image, wherein the annotation information includes a preset category associated with the first training image.

[0080] The preset classification model is iteratively trained using a first training data set until the classification model meets a preset convergence condition, thereby obtaining the first classification model.

[0081] In this embodiment, in order to be able to recognize the store identification information in the image to be tested based on the first classification model, it is necessary to firstly train the first classification model.

[0082] Optionally, a preset first training data set can be obtained, the first training data set including multiple first training data pairs, the first training data set including multiple first training data pairs, the first training data pair including a first training image and annotation information associated with the first training image, and the annotation information including a preset category associated with the first training image.

[0083] For example, the preset categories include a store image category, a store identification information category, and an irrelevant image category.

[0084] Furthermore, the preset classification model may be iteratively trained based on the first training data set until the classification model meets a preset convergence condition, thereby obtaining a first classification model.

[0085] For example, the training data pair can be input into the classification model to obtain the classification result output by the classification model. The current loss value of the classification model is determined based on the classification result and the annotation information corresponding to the training data. It is judged whether the loss value is less than the preset loss value threshold. If so, the current classification model can be determined as the second classification model. Otherwise, the parameters of the classification model can be adjusted in reverse based on the loss value until the loss value of the classification model is less than the preset loss value threshold, thereby obtaining the second classification model.

[0086] Among them, the preset convergence conditions include but are not limited to the loss value of the classification model being less than the preset loss value threshold, the difference between the loss values ​​of two rounds of training being less than the preset difference threshold, the training time reaching the preset time threshold, the number of iterations reaching the preset number threshold, etc.

[0087] By pre-setting preset categories such as store image categories, store identification information categories, and irrelevant image categories, and training the classification model based on the above preset categories, the first classification model obtained through training can output probability information that the image to be tested belongs to each preset category.

[0088] Furthermore, the preset category with the highest probability may be determined as the content category of the image to be tested.

[0089] In addition, if the content category of the image to be tested is determined to be an irrelevant image category, it means that the target area interval cannot be determined based on the image to be tested. Therefore, the image to be tested can be discarded.

[0090] The store area determination method provided in this embodiment pre-constructs a first training data set including a first training image and preset category annotation information associated with the first training image, and iteratively trains a preset classification model through the first training data set. Therefore, the trained first classification model can accurately implement the classification operation of the content category of the input image.

[0091] Figure 3 A flow chart of a method for determining a store area according to another embodiment of the present disclosure is provided. Based on any of the above embodiments, Figure 3 As shown, the method includes:

[0092] Step 301: Acquire an image to be tested associated with a target store, and determine a content category corresponding to the image to be tested by using a preset first classification model.

[0093] Step 302: If it is determined that the image to be tested is of a store logo type based on the content category, store logo information corresponding to the image to be tested is determined by a preset first object detection model.

[0094] Step 303: determine the store identification information as the target object.

[0095] Step 304: If it is determined that the image to be tested is a store image type based on the content category, store identification information of fast-moving consumer goods in the image to be tested is identified by a preset first object detection model.

[0096] Step 305: If the store identification information is recognized in the image to be tested, the store identification information is determined as the target object.

[0097] In this embodiment, after determining the content category corresponding to the image to be tested, different methods may be used to implement the recognition operation of the target object.

[0098] Optionally, if the image to be tested is determined to be a store identification type based on the content category, it indicates that the image to be tested only includes store identification information, so the store identification information corresponding to the image to be tested can be determined by a preset first target detection model. The store identification information is determined as the target object.

[0099] If the image to be tested is determined to be a store image type based on the content category, the image to be tested may be associated with target objects such as store logos, door openings, classification information, etc. Therefore, the detection operation of the target object can be achieved by combining the first target detection model, the second target detection model and the second classification model.

[0100] First, store identification information of fast-moving consumer goods in the image to be tested can be identified based on the preset first target detection model. If the store identification information is identified in the image to be tested, the store identification information is determined as the target object. On the contrary, if the store identification information is not identified in the image to be tested, the detection operation of the target object can be continued based on the combination of the second target detection model and the second classification model.

[0101] The store area determination method provided in this embodiment can accurately determine the target object associated with the image to be tested by performing a recognition operation of the store identification information through a preset first target detection model when determining that the image to be tested is of the store identification type. When the image to be tested is of the store image type, the store identification information contained in the image to be tested is recognized through a preset first target detection model, so that the target objects associated with the images to be tested of different content types can be accurately determined.

[0102] Further, based on any of the above embodiments, the method further includes:

[0103] A preset second training data set is obtained, where the second training data set includes a plurality of second training data pairs, where the second training data pairs include a second training image and annotation information associated with the second training image, and the annotation information includes store identification information associated with the second training image.

[0104] The preset target detection model is iteratively trained using the second training data set until the target detection model meets a preset convergence condition, thereby obtaining the first target detection model.

[0105] In this embodiment, in order to accurately identify the store identification information in the image to be tested, the first object detection model may be pre-trained.

[0106] Optionally, a preset second training data set may be obtained, the second training data set including a plurality of second training data pairs, the second training data pairs including a second training image and annotation information associated with the second training image, the annotation information including store identification information associated with the second training image.

[0107] For example, if the second training image is an image of a store under chain brand A, the corresponding annotation information may be chain brand A.

[0108] Furthermore, the preset target detection model can be iteratively trained using the second training data set until the target detection model meets a preset convergence condition, thereby obtaining a first target detection model.

[0109] For example, the second training data pair can be input into the target detection model to obtain the detection result output by the target detection model. The current loss value of the target detection model is determined based on the detection result and the annotation information corresponding to the training data. It is determined whether the loss value is less than the preset loss value threshold. If so, the current target detection model can be determined as the first target detection model. Otherwise, the parameters of the target detection model can be adjusted in reverse based on the loss value until the loss value of the target detection model is less than the preset loss value threshold, thereby obtaining the first target detection model.

[0110] Among them, the preset convergence conditions include but are not limited to the loss value of the classification model being less than the preset loss value threshold, the difference between the loss values ​​of two rounds of training being less than the preset difference threshold, the training time reaching the preset time threshold, the number of iterations reaching the preset number threshold, etc.

[0111] The store area determination method provided in the present embodiment pre-constructs a second training data set including multiple groups of second training images and store identification information associated with the second training images as annotation information, thereby iteratively training a preset target detection model based on the second training data set, so that the trained first target detection model can recognize the store identification information of the input image to be tested, and when store identification information exists, the target area range of the target store can be accurately determined based on the store identification information and the first mapping list.

[0112] Further, based on any of the above embodiments, after identifying the store identification information of the fast-moving consumer goods in the image to be tested by using the preset first object detection model, the method further includes:

[0113] If the store identification information is not recognized in the image to be tested, a preset second target detection model is used to identify a door opening quantity parameter associated with the image to be tested to obtain a recognition result.

[0114] If it is determined based on the recognition result that the image to be tested includes a door opening, the door opening quantity parameter is determined as the target object.

[0115] In this embodiment, the image to be tested may include door openings, wherein the door openings include but are not limited to display windows, entrance doors, etc. The area range of the store varies depending on the number of door openings. For example, if a store has one door opening, the corresponding area is less than 90 square meters. If a store has two door openings, the corresponding area range is 90-300 square meters.

[0116] Furthermore, if the image to be tested does not include store identification information, the door openings in the image to be tested may be identified to estimate the target area interval based on the door opening quantity parameter.

[0117] Optionally, a preset second target detection model can be used to identify the number of door openings associated with the image to be tested, and obtain a recognition result. The recognition result can be 1.0 door opening, indicating that there is a door opening in the image to be tested. Alternatively, the recognition result can also be 1.5 door openings, indicating that there is a regular-sized door opening and a smaller-sized door opening in the image to be tested. Alternatively, the recognition result can also be 0 door openings, indicating that there is no door opening in the image to be tested.

[0118] Further, if it is determined based on the recognition result that the image to be tested includes a door opening, the door opening quantity parameter is determined as the target object.

[0119] The store area determination method provided in this embodiment can accurately determine the target area range of the target store based on the door opening number parameter by identifying the door opening number parameter in the image to be tested through a preset second target detection model when it is determined that the image to be tested does not include store identification information.

[0120] Further, based on any of the above embodiments, before identifying the door opening quantity parameter associated with the image to be detected by using a preset second object detection model, the method further includes:

[0121] A preset third training data set is obtained, wherein the third training data set includes a plurality of third training data pairs, wherein the third training data pairs include a third training image and annotation information associated with the third training image, and the annotation information includes a parameter of the number of door openings in the third training image.

[0122] The preset target detection model is iteratively trained using the third training data set until the target detection model meets a preset convergence condition, thereby obtaining the second target detection model.

[0123] In this embodiment, in order to realize the recognition of the door opening in the image to be detected based on the second object detection model, the second object detection model may be trained in advance.

[0124] Optionally, a preset third training data set may be obtained, wherein the third training data set includes a plurality of third training data pairs, wherein the third training data pairs include a third training image and annotation information associated with the third training image, and the annotation information includes a parameter of the number of door openings in the third training image.

[0125] For example, if the third training image includes a door opening, the corresponding annotation information may be 1.0 door opening; if the third training image includes a smaller door opening, the corresponding annotation information may be 0.5 door opening; if the third training image does not include a door opening, the corresponding annotation information may be 0 door opening.

[0126] Furthermore, the preset target detection model can be iteratively trained using a third training data set until the target detection model meets a preset convergence condition, thereby obtaining the second target detection model.

[0127] For example, the third training data pair can be input into the target detection model to obtain the detection result output by the target detection model. The current loss value of the target detection model is determined based on the detection result and the annotation information corresponding to the training data. It is judged whether the loss value is less than the preset loss value threshold. If so, the current target detection model can be determined as the second target detection model. Otherwise, the parameters of the target detection model can be adjusted in reverse based on the loss value until the loss value of the target detection model is less than the preset loss value threshold, thereby obtaining the second target detection model.

[0128] Among them, the preset convergence conditions include but are not limited to the loss value of the classification model being less than the preset loss value threshold, the difference between the loss values ​​of two rounds of training being less than the preset difference threshold, the training time reaching the preset time threshold, the number of iterations reaching the preset number threshold, etc.

[0129] It should be noted that, since the annotation information associated with the third training image contains hidden information on the number of door openings, after the image to be tested is input into the second target detection model, the second target detection model can accurately detect whether the image to be tested contains door openings and the number of door openings.

[0130] The store area determination method provided in the present embodiment pre-constructs a third training data set including multiple groups of third training images and door opening number parameter annotation information associated with the third training images, thereby iteratively training a preset target detection model based on the third training data set, so that the trained second target detection model can accurately identify the door opening number parameters of the image to be tested, and further accurately determine the target area range of the target store based on the door opening number parameters.

[0131] Further, based on any of the above embodiments, if it is determined based on the recognition result that the image to be tested does not include a door opening, the image to be tested is classified by a preset second classification model to obtain classification information associated with the image to be tested, and the classification information is determined as the target object.

[0132] In this embodiment, when it is determined that the image to be tested does not include store identification information and door opening, the image to be tested is classified by the second classification model to determine the target area range of the target store based on the classification information.

[0133] Optionally, the image to be tested may be input into a preset second classification model, so that the second classification model can perform a classification operation on the image to be tested, obtain classification information associated with the image to be tested, and determine the classification information as a target object.

[0134] For example, the second classification model may output the probability that the image to be tested belongs to each preset category, and may determine the preset category with the highest probability as the classification information corresponding to the image to be tested.

[0135] The preset categories include, but are not limited to, large supermarkets, stores in shopping malls, wholesale stores, convenience stores, and irrelevant images. Different classification information corresponds to preset area intervals. For example, if the classification information of the target store is a large supermarket type, the corresponding area interval may be 1000-2000 square meters, and if the classification information of the target store is a store in a shopping mall, the corresponding area interval may be 500-1000 square meters. Therefore, after determining that the target object is classified information, the target area interval associated with the target store can be accurately determined based on the classification information.

[0136] The store area determination method provided in this embodiment, when the image to be tested does not include store identification information and door opening, classifies the image to be tested through a second classification model, so that the target area range of the target store can be accurately determined based on the classification information corresponding to the image to be tested.

[0137] Further, based on any of the above embodiments, before the classification operation is performed on the image to be tested by using the preset second classification model, it also includes:

[0138] A preset fourth training data set is obtained, wherein the fourth training data set includes a plurality of fourth training data pairs, wherein the fourth training data pairs include a fourth training image and annotation information associated with the fourth training image, and the annotation information is classification information associated with the fourth training image.

[0139] The preset classification model is iteratively trained using the fourth training data set until the classification model meets a preset convergence condition, thereby obtaining the second classification model.

[0140] In this embodiment, in order to perform a classification operation on the image to be tested based on the second classification model, the second classification model needs to be trained first.

[0141] Optionally, a preset fourth training data set may be obtained, wherein the fourth training data set includes a plurality of fourth training data pairs, wherein the fourth training data pair includes a fourth training image and annotation information associated with the fourth training image, and the annotation information is classification information associated with the fourth training image. The classification information includes, but is not limited to, large supermarkets, stores in shopping malls, wholesale stores, convenience stores, and irrelevant images.

[0142] Furthermore, the preset classification model may be iteratively trained based on the fourth training data set until the classification model meets a preset convergence condition, thereby obtaining a second classification model.

[0143] For example, the training data pair can be input into the classification model to obtain the classification result output by the classification model. The current loss value of the classification model is determined based on the classification result and the annotation information corresponding to the training data. It is judged whether the loss value is less than the preset loss value threshold. If so, the current classification model can be determined as the second classification model. Otherwise, the parameters of the classification model can be adjusted in reverse based on the loss value until the loss value of the classification model is less than the preset loss value threshold, thereby obtaining the second classification model.

[0144] Among them, the preset convergence conditions include but are not limited to the loss value of the classification model being less than the preset loss value threshold, the difference between the loss values ​​of two rounds of training being less than the preset difference threshold, the training time reaching the preset time threshold, the number of iterations reaching the preset number threshold, etc.

[0145] The store area determination method provided in the present embodiment uses the store area determination method provided in the present embodiment, and uses the fourth training image and the classification information associated with the fourth training image as the fourth training data set as the fourth training data set for labeling information, so that the preset classification model can be iteratively trained based on the fourth training data set, so that the trained second classification model can accurately perform recognition operations on the classification information of the image to be tested.

[0146] Optionally, based on any of the above embodiments, step 203 includes:

[0147] If the target object is store identification information, a preset first mapping list is determined as the target mapping list, wherein the first mapping list includes mapping relationships between different store identification information and area intervals.

[0148] If the target object is a door opening quantity parameter, a preset second mapping list is determined as the target mapping list, wherein the second mapping list includes mapping relationships between different door opening quantity parameters and area intervals.

[0149] If the target object is classification information, a preset third mapping list is determined as the target mapping list, and the third mapping list includes mapping relationships between different classification information and area intervals.

[0150] In this embodiment, in order to achieve accurate estimation of the area range of the target store, a mapping list corresponding to multiple different target objects may be pre-constructed.

[0151] Optionally, the mapping list may include a first mapping list, a second mapping list, and a third mapping list. Among them, the first mapping list includes a mapping relationship between different store identification information and area intervals. For example, the area interval corresponding to the stores under the A chain brand is 80-90 square meters. The area interval corresponding to the stores under the B chain brand is 90-100 square meters. The second mapping list includes a mapping relationship between different doorway quantity parameters and area intervals. For example, if there is one doorway in the store, the corresponding area is less than 90 square meters. If there are two doorways in the store, the corresponding area interval is 90-300 square meters. The third mapping list includes a mapping relationship between different classification information and area intervals. For example, if the classification information of the target store is a large supermarket type, the corresponding area interval can be 1000-2000 square meters. If the classification information of the target store is a store in a shopping mall, the corresponding area interval can be 500-1000 square meters.

[0152] Therefore, after determining the target object associated with the image to be tested, a target mapping list matching the target object can be determined. If the target object is store identification information, the preset first mapping list is determined as the target mapping list; if the target object is a door opening quantity parameter, the preset second mapping list is determined as the target mapping list; if the target object is classification information, the preset third mapping list is determined as the target mapping list.

[0153] The store area determination method provided in this embodiment pre-builds a mapping list associated with multiple target objects, so that after identifying the target object associated with the image to be tested, the target area range of the target store can be accurately determined based on the mapping list.

[0154] Further, based on any of the above embodiments, step 204 includes:

[0155] An area interval matching the target object is determined in the target mapping list.

[0156] An area interval matching the target object is determined as the target area interval.

[0157] In this embodiment, after the target object associated with the image to be tested is identified and the target mapping list corresponding to the target object is determined, the target mapping list can be traversed to determine the target object in the target mapping list. The area interval corresponding to the target object in the target mapping list is determined, and the area interval matching the target object is determined as the target area interval.

[0158] For example, if the image to be tested includes store identification information, based on the store identification information, it can be determined that the target store is a store under chain brand A. The store area interval corresponding to chain brand A can be determined in the first mapping list. The area interval is determined as the target area interval of the target store.

[0159] Still taking a practical application as an example, if it is determined that the image to be tested includes 2.0 door openings, the area interval matching the 2.0 door openings can be determined in the preset second mapping list. The area interval is determined as the target area interval of the target store. For example, if the store has one door opening, the corresponding area is less than 90 square meters. If the store has two door openings, the corresponding area interval is 90-300 square meters.

[0160] Figure 4 A schematic diagram of an application scenario provided by an embodiment of the present disclosure, such as Figure 4 As shown, the image to be tested 41 associated with the target store can be obtained, and the content category associated with the image to be tested 41 can be determined by the preset first classification model 42. If the image to be tested is determined to be a store identification type by the content category, the store identification information corresponding to the image to be tested 41 is determined by the preset first target detection model 43. The target area interval 44 of the target store is determined according to the preset first mapping list and the store identification information. If the image to be tested 41 is determined to be a store image type based on the content category, the store identification information of the fast-moving consumer goods in the image to be tested is identified by the preset first target detection model 43. If the image to be tested 41 includes store identification information, the target area interval 44 of the target store is determined based on the preset first mapping list 45 and the store identification information. If the image to be tested 41 does not include store identification information, the door opening number parameter associated with the image to be tested 41 is further identified based on the preset second target detection model 46. If it is determined that the image to be tested 41 includes a door opening based on the door opening number parameter, the target area interval 44 of the target store is determined based on the preset second mapping list 47 and the door opening number parameter. On the contrary, if the image to be tested 41 does not include a door opening, the image to be tested 41 is classified based on the preset second classification model 48 to obtain classification information associated with the image to be tested 41, and the target area range 44 of the target store is determined based on the classification information and the preset third mapping list 49.

[0161] The store area determination method provided in this embodiment can determine the target area interval of the target store by determining the area interval associated with the target object in the target mapping list after respectively determining the target object and the target mapping list.

[0162] Figure 5 A flow chart of a method for determining store area provided in another embodiment of the present disclosure, based on any of the above embodiments, the method further includes:

[0163] Step 501: If the store identification information is not recognized in the image to be tested, the image to be tested is input into the second target detection model and the second classification model respectively to obtain the detection result output by the second target detection model and the target classification result output by the second classification model.

[0164] Step 502: If the area interval determined based on the detection result is different from the area interval determined based on the target classification result, the area interval determined based on the detection result is determined as the target area interval.

[0165] In this embodiment, when it is determined that the image to be tested is a store image type and the image to be tested does not include store identification information, the image to be tested can be input into the second target detection model and the second classification model respectively to obtain the detection result output by the second target detection model and the target classification result output by the second classification model.

[0166] After the detection result and the target classification result are determined respectively, the target area interval corresponding to the target store can be determined based on the detection result and the second mapping list. In addition, the target area interval corresponding to the target store can also be determined based on the target classification result and the third mapping list. If the area interval determined based on the detection result is different from the area interval determined based on the target classification result, the area interval determined based on the detection result is determined as the target area interval.

[0167] Taking practical application as an example, a comprehensive supermarket can be opened in a large shopping mall. Generally speaking, a comprehensive supermarket opened in a large shopping mall generally does not have external door openings. After the image to be tested is input into the second target detection model and the second classification model respectively, if the second target detection model detects that the target store includes a door opening, but the second classification model outputs that the target store is a comprehensive supermarket in a large shopping mall. Then there are differences in the stores determined by the two models. After determining the area interval based on the detection results and determining the area interval based on the target classification results, in order to improve the accuracy of the target area interval, the area interval determined based on the detection results can be determined as the target area interval.

[0168] The store area determination method provided in this embodiment inputs the image to be tested into the second target detection model and the second classification model at the same time when the image to be tested does not include store identification information, so that the output results of the two models can be mutually verified to improve the accuracy of area determination.

[0169] Figure 6 A schematic diagram of the structure of a store area determination device provided in an embodiment of the present disclosure, such as Figure 6 As shown, the device includes: an acquisition module 61, a determination module 62, a mapping module 63 and a processing module 64. Among them, the acquisition module 61 is used to acquire the image to be tested associated with the target store, and determine the content category corresponding to the image to be tested by a preset first classification model. The determination module 62 is used to determine the target object associated with the image to be tested by using a target detection model matching the content category. The mapping module 63 is used to determine a target mapping list matching the target object, the target mapping list includes multiple mapping relationship pairs, and the mapping relationship pairs include the target object and the area interval corresponding to the target object. The processing module 64 is used to determine the target area interval corresponding to the target store based on the target mapping list and the target object.

[0170] Further, based on any of the above embodiments, the acquisition module includes: a processing unit, configured to input the image to be tested into a preset first classification model, and obtain probability information output by the first classification model, wherein the probability information includes prediction probabilities corresponding to a plurality of preset categories; and a determination unit, configured to determine a preset category whose prediction probability meets a preset condition as a content category corresponding to the image to be tested.

[0171] Further, based on any of the above embodiments, the device further includes: an acquisition module for acquiring a first training data set, the first training data set including a plurality of first training data pairs, the first training data pairs including a first training image and annotation information associated with the first training image, the annotation information including a preset category associated with the first training image. A training module for iteratively training a preset classification model using the first training data set until the classification model meets a preset convergence condition, thereby obtaining the first classification model.

[0172] Further, based on any of the above embodiments, the determination module includes: an identification unit, which is used to determine the store identification information corresponding to the image to be tested by a preset first target detection model if the image to be tested is determined to be a store identification type based on the content category. A determination unit, which is used to determine the store identification information as the target object.

[0173] Further, based on any of the above embodiments, the determination module includes: an identification unit, which is used to identify the store identification information of the fast-moving consumer goods in the image to be tested by a preset first target detection model if the image to be tested is determined to be a store image type based on the content category. A determination unit, which is used to determine the store identification information as the target object if the store identification information is identified in the image to be tested.

[0174] Further, based on any of the above embodiments, the device further includes: an acquisition module for acquiring a preset second training data set, wherein the second training data set includes a plurality of second training data pairs, wherein the second training data pairs include a second training image and annotation information associated with the second training image, wherein the annotation information includes store identification information associated with the second training image. A training module for iteratively training a preset target detection model using the second training data set until the target detection model meets a preset convergence condition, thereby obtaining the first target detection model.

[0175] Further, based on any of the above embodiments, the determination module further includes: an identification unit, for identifying the door opening quantity parameter associated with the image to be tested by a preset second target detection model to obtain a recognition result if the store identification information is not identified in the image to be tested; and a determination unit, for determining the door opening quantity parameter as the target object if it is determined based on the recognition result that the image to be tested includes a door opening.

[0176] Further, based on any of the above embodiments, the device further includes: an acquisition module for acquiring a preset third training data set, wherein the third training data set includes a plurality of third training data pairs, wherein the third training data pairs include a third training image and annotation information associated with the third training image, wherein the annotation information includes a parameter of the number of door openings in the third training image. A training module for iteratively training a preset target detection model using the third training data set until the target detection model meets a preset convergence condition, thereby obtaining the second target detection model.

[0177] Furthermore, based on any of the above embodiments, the determination module also includes: a classification unit, which is used to classify the image to be tested through a preset second classification model if it is determined based on the recognition result that the image to be tested does not include a door opening, obtain classification information associated with the image to be tested, and determine the classification information as the target object.

[0178] Further, based on any of the above embodiments, the device further includes: an acquisition module, used to acquire a preset fourth training data set, wherein the fourth training data set includes a plurality of fourth training data pairs, wherein the fourth training data pairs include a fourth training image and annotation information associated with the fourth training image, wherein the annotation information is classification information associated with the fourth training image. A training module, used to iteratively train a preset classification model using the fourth training data set until the classification model meets a preset convergence condition, thereby obtaining the second classification model.

[0179] Further, based on any of the above embodiments, the device further includes: a classification module, which is used to input the image to be tested into the second target detection model and the second classification model respectively if the store identification information is not recognized in the image to be tested, and obtain the detection result output by the second target detection model and the target classification result output by the second classification model. A determination module, which is used to determine the area interval determined based on the detection result as the target area interval if the area interval determined based on the detection result is different from the area interval determined based on the target classification result.

[0180] Further, on the basis of any of the above embodiments, the mapping module includes: a first determination unit, for determining a preset first mapping list as the target mapping list if the target object is store identification information, the first mapping list including mapping relationships between different store identification information and area intervals. A second determination unit, for determining a preset second mapping list as the target mapping list if the target object is a door opening quantity parameter, the second mapping list including mapping relationships between different door opening quantity parameters and area intervals. A third determination unit, for determining a preset third mapping list as the target mapping list if the target object is classification information, the third mapping list including mapping relationships between different classification information and area intervals.

[0181] Further, based on any of the above embodiments, the processing module includes: a determination unit, configured to determine an area interval matching the target object in the target mapping list; and a processing unit, configured to determine the area interval matching the target object as the target area interval.

[0182] Further, based on any of the above embodiments, the device further comprises: a storage module, used to associate and store the identification information of the target store and the target area interval in a preset storage path. Alternatively, an update module, used to perform a data update operation on the associated parameters of the target store in the map software based on the target area interval associated with the target store.

[0183] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0184] According to an embodiment of the present disclosure, the present disclosure further provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to execute the method described in any one of the above embodiments.

[0185] According to an embodiment of the present disclosure, the present disclosure further provides a computer program product, including a computer program, which implements the steps of the method described in any of the above embodiments when executed by a processor.

[0186] According to an embodiment of the present disclosure, the present disclosure also provides a computer program product, which includes: a computer program, the computer program is stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program so that the electronic device executes the solution provided by any of the above embodiments.

[0187] Figure 7 A schematic diagram of the structure of an electronic device provided for an embodiment of the present disclosure. The electronic device 700 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0188] like Figure 7 As shown, the device 700 includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0189] A number of components in the device 700 are connected to the I / O interface 705, including: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a disk, an optical disk, etc.; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0190] The computing unit 701 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 701 performs the various methods and processes described above, such as the store area determination method. For example, in some embodiments, the store area determination method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 700 via the ROM 702 and / or the communication unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the store area determination method described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform the store area determination method in any other appropriate manner (e.g., by means of firmware).

[0191] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0192] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0193] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0194] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0195] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0196] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0197] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0198] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for determining a store area, comprising: Acquire the image to be tested associated with the target store, and determine the content category corresponding to the image to be tested by using a preset first classification model; Determine the target object associated with the image to be detected by using a target detection model that matches the content category; Determine a target mapping list matching the target object, wherein the target mapping list includes a plurality of mapping relationship pairs, and the mapping relationship pairs include the target object and an area interval corresponding to the target object; A target area interval corresponding to the target store is determined based on the target mapping list and the target object.

2. The method according to claim 1, wherein: The determining the content category corresponding to the image to be tested by using a preset first classification model includes: Inputting the image to be tested into a preset first classification model to obtain probability information output by the first classification model, wherein the probability information includes prediction probabilities corresponding to a plurality of preset categories; The preset category whose predicted probability meets the preset condition is determined as the content category corresponding to the image to be tested.

3. The method according to claim 2, before determining the content category corresponding to the image to be tested by using a preset first classification model, further comprising: Acquire a first training data set, the first training data set including a plurality of first training data pairs, the first training data pairs including a first training image and annotation information associated with the first training image, the annotation information including a preset category associated with the first training image; The preset classification model is iteratively trained using a first training data set until the classification model meets a preset convergence condition, thereby obtaining the first classification model.

4. The method according to claim 1, wherein: The step of using a target detection model that matches the content category to determine a target object associated with the image to be detected includes: If the image to be tested is determined to be a store logo type based on the content category, determining the store logo information corresponding to the image to be tested by using a preset first object detection model; The store identification information is determined as the target object.

5. The method according to claim 1, wherein: The step of using a target detection model that matches the content category to determine a target object associated with the image to be detected includes: If it is determined that the image to be tested is a store image type based on the content category, identifying the store identification information of the fast-moving consumer goods in the image to be tested by using a preset first object detection model; If the store identification information is recognized in the image to be tested, the store identification information is determined as the target object.

6. The method according to claim 4 or 5, further comprising: Acquire a preset second training data set, where the second training data set includes a plurality of second training data pairs, where the second training data pairs include a second training image and annotation information associated with the second training image, where the annotation information includes store identification information associated with the second training image; The preset target detection model is iteratively trained using the second training data set until the target detection model meets a preset convergence condition, thereby obtaining the first target detection model.

7. The method according to claim 5, after identifying the store identification information of the fast-moving consumer goods in the image to be tested by the preset first object detection model, further comprising: If the store identification information is not recognized in the image to be tested, a door opening quantity parameter associated with the image to be tested is recognized by a preset second object detection model to obtain a recognition result; If it is determined based on the recognition result that the image to be tested includes a door opening, the door opening quantity parameter is determined as the target object.

8. The method according to claim 7, before identifying the door opening quantity parameter associated with the image to be detected by using a preset second target detection model, further comprising: Acquire a preset third training data set, wherein the third training data set includes a plurality of third training data pairs, wherein the third training data pairs include a third training image and annotation information associated with the third training image, wherein the annotation information includes a parameter of the number of door openings in the third training image; The preset target detection model is iteratively trained using the third training data set until the target detection model meets a preset convergence condition, thereby obtaining the second target detection model.

9. The method according to claim 7, after identifying the door opening quantity parameter associated with the image to be detected by using the preset second target detection model, further comprising: If it is determined based on the recognition result that the image to be tested does not include a door opening, a classification operation is performed on the image to be tested using a preset second classification model to obtain classification information associated with the image to be tested, and the classification information is determined as the target object.

10. The method according to claim 9, before classifying the image to be tested by using a preset second classification model, further comprising: Acquire a preset fourth training data set, wherein the fourth training data set includes a plurality of fourth training data pairs, wherein the fourth training data pairs include a fourth training image and annotation information associated with the fourth training image, wherein the annotation information is classification information associated with the fourth training image; The preset classification model is iteratively trained using the fourth training data set until the classification model meets a preset convergence condition, thereby obtaining the second classification model.

11. The method according to claim 5, further comprising: If the store identification information is not recognized in the image to be tested, the image to be tested is input into the second target detection model and the second classification model respectively to obtain the detection result output by the second target detection model and the target classification result output by the second classification model; If the area interval determined based on the detection result is different from the area interval determined based on the target classification result, the area interval determined based on the detection result is determined as the target area interval.

12. The method according to any one of claims 1 to 5 or 7 to 11, wherein: The determining a target mapping list matching the target object includes: If the target object is store identification information, a preset first mapping list is determined as the target mapping list, wherein the first mapping list includes mapping relationships between different store identification information and area intervals; If the target object is a door opening quantity parameter, a preset second mapping list is determined as the target mapping list, wherein the second mapping list includes mapping relationships between different door opening quantity parameters and area intervals; If the target object is classification information, a preset third mapping list is determined as the target mapping list, and the third mapping list includes mapping relationships between different classification information and area intervals.

13. The method according to any one of claims 1 to 5 or 7 to 11, wherein: The determining the target area interval corresponding to the target store based on the target mapping list and the target object includes: Determining an area interval matching the target object in the target mapping list; An area interval matching the target object is determined as the target area interval.

14. The method according to any one of claims 1-5 or 7-11, after determining the target area interval corresponding to the target store based on the target mapping list and the target object, further comprising: The identification information of the target store and the target area interval are associated and stored in a preset storage path; or, Based on the target area interval associated with the target store, data updating operation is performed on the associated parameters of the target store in the map software.

15. A device for determining store area, comprising: An acquisition module, used to acquire an image to be tested associated with a target store, and determine a content category corresponding to the image to be tested by using a preset first classification model; A determination module, configured to determine a target object associated with the image to be detected by using a target detection model that matches the content category; A mapping module, used to determine a target mapping list matching the target object, wherein the target mapping list includes a plurality of mapping relationship pairs, and the mapping relationship pairs include the target object and an area interval corresponding to the target object; A processing module is used to determine a target area interval corresponding to the target store based on the target mapping list and the target object.

16. The device according to claim 15, wherein: The acquisition module comprises: A processing unit, configured to input the image to be tested into a preset first classification model, and obtain probability information output by the first classification model, wherein the probability information includes prediction probabilities corresponding to a plurality of preset categories; A determination unit is used to determine a preset category whose prediction probability meets a preset condition as the content category corresponding to the image to be tested.

17. The apparatus according to claim 16, further comprising: an acquisition module, configured to acquire a first training data set, wherein the first training data set includes a plurality of first training data pairs, wherein the first training data pairs include a first training image and annotation information associated with the first training image, wherein the annotation information includes a preset category associated with the first training image; The training module is used to iteratively train a preset classification model using a first training data set until the classification model meets a preset convergence condition, thereby obtaining the first classification model.

18. The device according to claim 15, wherein: The determination module comprises: an identification unit, configured to determine the store identification information corresponding to the image to be tested by using a preset first object detection model if the image to be tested is determined to be a store identification type based on the content category; A determination unit is used to determine the store identification information as the target object.

19. The device according to claim 15, wherein: The determination module comprises: an identification unit, configured to identify store identification information of fast-moving consumer goods in the image to be tested by using a preset first object detection model if the image to be tested is determined to be a store image type based on the content category; A determination unit is configured to determine the store identification information as the target object if the store identification information is recognized in the image to be tested.

20. The device according to claim 18 or 19, further comprising: an acquisition module, configured to acquire a preset second training data set, wherein the second training data set includes a plurality of second training data pairs, wherein the second training data pairs include a second training image and annotation information associated with the second training image, wherein the annotation information includes store identification information associated with the second training image; The training module is used to iteratively train the preset target detection model through the second training data set until the target detection model meets the preset convergence condition, thereby obtaining the first target detection model.

21. The apparatus according to claim 19, wherein the determining module further comprises: an identification unit, configured to identify the door opening quantity parameter associated with the image to be tested by using a preset second target detection model to obtain an identification result if the store identification information is not identified in the image to be tested; A determination unit is configured to determine a door opening quantity parameter as the target object if it is determined based on the recognition result that the image to be tested includes a door opening.

22. The apparatus according to claim 21, further comprising: an acquisition module, configured to acquire a preset third training data set, wherein the third training data set includes a plurality of third training data pairs, wherein the third training data pairs include a third training image and annotation information associated with the third training image, wherein the annotation information includes a parameter of the number of door openings in the third training image; The training module is used to iteratively train the preset target detection model through a third training data set until the target detection model meets a preset convergence condition, thereby obtaining the second target detection model.

23. The apparatus according to claim 21, wherein the determining module further comprises: A classification unit is used to classify the image to be tested by using a preset second classification model if it is determined based on the recognition result that the image to be tested does not include a door opening, obtain classification information associated with the image to be tested, and determine the classification information as the target object.

24. The apparatus according to claim 23, further comprising: an acquisition module, configured to acquire a preset fourth training data set, wherein the fourth training data set includes a plurality of fourth training data pairs, wherein the fourth training data pairs include a fourth training image and annotation information associated with the fourth training image, wherein the annotation information is classification information associated with the fourth training image; The training module is used to iteratively train the preset classification model through a fourth training data set until the classification model meets a preset convergence condition, thereby obtaining the second classification model.

25. The apparatus according to claim 19, further comprising: A classification module, configured to input the image to be tested into the second target detection model and the second classification model respectively, if the store identification information is not recognized in the image to be tested, to obtain a detection result output by the second target detection model and a target classification result output by the second classification model; A determination module is configured to determine the area interval determined based on the detection result as the target area interval if the area interval determined based on the detection result is different from the area interval determined based on the target classification result.

26. The device according to any one of claims 15 to 29 or 21 to 25, wherein: The mapping module comprises: A first determining unit, configured to determine a preset first mapping list as the target mapping list if the target object is store identification information, wherein the first mapping list includes mapping relationships between different store identification information and area intervals; a second determining unit, configured to determine a preset second mapping list as the target mapping list if the target object is a door opening quantity parameter, wherein the second mapping list includes mapping relationships between different door opening quantity parameters and area intervals; The third determining unit is configured to determine a preset third mapping list as the target mapping list if the target object is classification information, wherein the third mapping list includes mapping relationships between different classification information and area intervals.

27. The device according to any one of claims 15 to 29 or 21 to 25, wherein: The processing module comprises: A determination unit, configured to determine an area interval matching the target object in the target mapping list; A processing unit is used to determine an area interval matching the target object as the target area interval.

28. The device according to any one of claims 15-29 or 21-25, further comprising: A storage module, used for associating and storing the identification information of the target store and the target area interval in a preset storage path; or, An updating module is used to perform data updating operations on the associated parameters of the target store in the map software based on the target area interval associated with the target store.

29. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 14.

30. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-14.

31. A computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 14.

Images