WO2023077940A1

WO2023077940A1 - Recognition method and apparatus, processing method and apparatus, and intelligent detection system

Info

Publication number: WO2023077940A1
Application number: PCT/CN2022/116767
Authority: WO
Inventors: 田俊文; 王元元; 崔晓静; 孔滕滕; 赵贤宇
Original assignee: 北京沃东天骏信息技术有限公司; 北京京东世纪贸易有限公司
Priority date: 2021-11-03
Filing date: 2022-09-02
Publication date: 2023-05-11
Also published as: CN113903033A

Abstract

The present application relates to the technical field of data processing. Disclosed are a recognition method and apparatus, a processing method and apparatus, and an intelligent detection system. A specific embodiment of the recognition method comprises: acquiring image information and authorized brand information of a target item; on the basis of the image information, determining whether the brand of the target item is a preset brand; in response to determining that the brand of the target item is the preset brand, matching the authorized brand information of the target item with brand information of the preset brand; and on the basis of a matching result, determining a recognition result of the target item. By means of the embodiment, the recognition accuracy and efficiency are effectively improved.

Description

Identification, processing method and device, and intelligent detection system

Cross References to Related Applications

This patent application claims the priority of the Chinese patent application with the application number 202111295792.6 and the title of the invention "identification, processing method and device and intelligent detection system" filed on November 3, 2021, the full text of which is incorporated by reference incorporated into this application.

technical field

The present application relates to the field of computer technology, specifically to the field of data processing technology, and in particular to an identification and processing method and device and an intelligent detection system.

Background technique

With the development of society and the improvement of policies and regulations, the protection of intellectual property rights has been paid more and more attention in various fields. The protection of intellectual property rights is of great significance to the progress of science and technology and the healthy and orderly development of the market economy. The intellectual property rights of core technologies are even more important for enterprises. assets.

Contents of the invention

Embodiments of the present application provide an identification method, device, equipment, and storage medium.

According to the first aspect, an embodiment of the present application provides an identification method, the method comprising: acquiring image information and authorized brand information of a target item; determining whether the brand of the target item is a preset brand based on the image information; The brand of the item is a preset brand, and the authorized brand information of the target item is matched with the brand information of the preset brand; based on the matching result, the recognition result of the target item is determined.

According to the second aspect, the embodiment of the present application provides a processing method, the method includes: obtaining the identification result of the item to be processed, and the identification result is obtained by identifying the item to be processed by using the method described in any implementation manner of the first aspect above The recognition result; based on the recognition result, the processing operation is performed on the item to be processed.

According to a third aspect, an embodiment of the present application provides an identification device, which includes: an acquisition module configured to acquire image information and authorized brand information of a target item; a determination module configured to determine the target item based on the image information Whether the brand of the target item is a preset brand; the matching module is configured to match the authorized brand information of the target item with the brand information of the preset brand in response to determining that the brand of the target item is a preset brand; the identification module is configured to be based on The matching result determines the recognition result of the target item.

According to a fourth aspect, an embodiment of the present application provides a processing device, which includes: a processing module configured to obtain the identification result of the item to be processed, and the identification result is the method described in any implementation manner of the above-mentioned first aspect The identification result obtained by identifying the item to be processed; the execution module is configured to perform a processing operation on the item to be processed based on the identification result.

According to a fifth aspect, an embodiment of the present application provides a smart inspection system, the system includes: a service module, configured to apply a big data tool to obtain item data of a target item from a database, and send the image information of the target item to the detection module To detect whether the brand of the target item is a preset brand and obtain a detection result, and in response to determining that the brand of the target item is a default brand, generate an identification result of the target item based on the brand information of the preset brand and the authorized brand information, and The recognition result is stored in the cloud storage module; the detection module is used to detect whether the brand of the target item is a preset brand based on the image information of the target item, and sends the detection result to the service module; the cloud storage module is used to identify the target item. The identification result is stored; the audit system back-end module is used to obtain the identification result of the target item from the cloud storage module, and regularly pushes the identification result to the external governance module; the external processing module is used to process the target item according to the identification result of the target item The target item performs preset processing operations, and generates the final confirmation result of the target item.

According to a sixth aspect, an embodiment of the present application provides an electronic device, the electronic device includes one or more processors; a storage device, on which one or more programs are stored, when the one or more programs are Multiple processors are executed, so that one or more processors implement the method described in any implementation manner of the first aspect or the second aspect.

According to a seventh aspect, an embodiment of the present application provides a computer-readable medium, on which a computer program is stored. When the program is executed by a processor, the method described in any implementation manner of the first aspect or the second aspect is implemented.

It should be understood that what is described in this section is not intended to identify key or important features of the embodiments of the application, nor is it intended to limit the scope of the application. Other features of the present application will be easily understood from the following description.

Description of drawings

FIG. 1 is an exemplary system architecture diagram to which the present application can be applied;

Fig. 2 is a flowchart according to an embodiment of the identification method of the present application;

Fig. 3 is a schematic diagram of an application scenario of the recognition method according to the present application;

Fig. 4 is a flowchart according to another embodiment of the identification method of the present application;

Fig. 5 is a flow chart according to one embodiment of the processing method of the present application;

Fig. 6 is a schematic diagram of an embodiment of an identification device according to the present application;

Figure 7 is a schematic diagram of one embodiment of a processing device according to the present application;

Fig. 8 is a schematic diagram according to an embodiment of the intelligent inspection system of the present application;

FIG. 9 is a schematic structural diagram of a computer system suitable for implementing the server of the embodiment of the present application.

Detailed ways

Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and they should be regarded as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and embodiments.

FIG. 1 shows an exemplary system architecture 100 to which embodiments of the recognition method of the present application can be applied.

As shown in FIG. 1 , a system architecture 100 may include

terminal devices

101 , 102 , 103 , a network 104 and a server 105 . The network 104 is used as a medium for providing communication links between the

terminal devices

101 , 102 , 103 and the server 105 . Network 104 may include various connection types, such as wires, wireless communication links, or fiber optic cables, among others.

The

terminal devices

101, 102, 103 interact with the server 105 via the network 104 to receive or send messages and the like. Various communication client applications can be installed on the

terminal devices

101, 102, 103, such as shopping applications, communication applications, and the like.

The

terminal devices

101, 102, and 103 may be hardware or software. When the

terminal devices

101, 102, 103 are hardware, they may be various electronic devices with display screens, including but not limited to mobile phones and notebook computers. When the

terminal devices

101, 102, 103 are software, they can be installed in the electronic devices listed above. It can be implemented as a plurality of software or software modules (for example, to provide identification services), or as a single software or software module. No specific limitation is made here.

The server 105 may be a server providing various services, for example, acquiring image information and authorized brand information of the target item; determining whether the brand of the target item is a preset brand based on the image information; , matching the authorized brand information of the target item with the brand information of the preset brand; and determining the recognition result of the target item based on the matching result.

It should be noted that the server 105 may be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, for providing identification services), or as a single software or software module. No specific limitation is made here.

It should be pointed out that the identification method provided by the embodiment of the present application may be executed by the server 105, or executed by the

terminal devices

101, 102, 103, or executed by the server 105 and the

terminal devices

101, 102, 103 in cooperation with each other. Correspondingly, each part (such as each unit, subunit, module, submodule) included in the identification device can be all set in the server 105, can also be all set in the

terminal equipment

101, 102, 103, can also be set in the server respectively 105 and

terminal devices

101, 102, 103.

It should be understood that the numbers of terminal devices, networks and servers in Fig. 1 are only illustrative. According to the implementation needs, there can be any number of terminal devices, networks and servers.

FIG. 2 shows a schematic flowchart 200 of an identification method that can be applied to the present application. In this embodiment, the identification method includes the following steps:

Step 201, acquire the image information and authorized brand information of the target item.

In this embodiment, the execution subject (server 105 or

terminal devices

101, 102, 103 as shown in FIG. 1 ) can actively obtain the image information and authorized brand information of the target item to be identified from a local or remote server, or can The image information and authorized brand information of the target item to be identified specified by the user are passively received, which is not limited in this application.

Wherein, the target item may be any item to be identified, for example, clothing, lamps and the like.

Here, the image information of the target item is used to indicate the image containing the target item and its associated information (for example, title information, corresponding store information, etc.). It should be pointed out that the image containing the target item may only contain the image of the target item, or It may contain not only the image of the target item, but also images of other items except the target item, which is not limited in this application.

The authorized brand information of the target item is used to indicate the authorized brand information of the target item, which can be determined through the store information corresponding to the target item, for example, store ID, store operating qualification information, etc.

Step 202, based on the image information, determine whether the brand of the target item is a preset brand.

In this embodiment, the execution subject may determine whether the brand of the target item is a preset brand according to the image information of the target item.

Wherein, the brand information of the target item is used to indicate all information related to the brand of the target item received by consumers and other shareholders, for example, brand name, authorized store ID (Identity document, identity identification number) and so on.

Here, the execution subject can directly determine whether the brand of the target item is a preset brand according to the image of the item included in the image information of the target item, and can determine the brand name of the target item according to the logo of the target item included in the image information of the target item. Whether the brand is a default brand or not can also be determined according to the style of the target item included in the image information of the target item, and this application does not limit this.

Step 203, in response to determining that the brand of the target item is a preset brand, match the authorized brand information of the target item with the brand information of the preset brand.

In this embodiment, the execution subject can determine whether the brand of the target item is the default brand according to the image information of the target item, and if it is determined that the brand of the target item is the default brand, then compare the authorized brand information of the target item with the preset brand brand information for matching.

Step 204, based on the matching result, determine the recognition result of the target item.

In this embodiment, after the execution subject obtains the matching result, it can judge the matching result. If the matching result is a matching failure, it can be determined that the recognition result of the target item is an item that meets the preset conditions; if the matching result is a matching If it succeeds, it can be determined that the recognition result of the target item is an item that does not meet the preset condition.

Here, the preset conditions can be set according to actual needs, for example, whether the item complies with regulations, whether it is infringing, etc., which is not limited in this application.

Specifically, the preset condition is set according to whether the item is compliant. If the matching result is a matching failure, it can be determined that the recognition result of the target item is an item that meets the preset condition, that is, an unqualified item; if the matching result is a matching success, Then it can be determined that the identification result of the target item is an item that does not meet the preset condition, that is, a compliant item.

In some optional manners, determining the identification result of the target item based on the matching result includes: determining that the target item is an infringing item in response to determining that the matching result is a matching failure.

In this implementation mode, after obtaining the matching result, the execution subject can judge the matching result. If the matching result is a successful match, it will identify that the target item is not an infringing item, that is, identify the target item as an item with a preset brand authorization. If the matching result is a matching failure, the identification target item is an infringing item, that is, the identification target item is an item that does not have a preset brand authorization.

This implementation manner helps to improve the accuracy and efficiency of identifying infringing items by determining that the target item is an infringing item in response to determining that the matching result is a matching failure.

In some optional manners, based on the matching result, determining the recognition result of the target item includes: in response to determining that the matching result is a matching failure, obtaining the user's recognition result of the target item; based on the user's recognition result of the target item, determining the target Item recognition results.

In this implementation, after obtaining the matching result, the execution subject can judge the matching result. If the matching result is a matching failure, the image information of the target item, the authorized brand information and the brand information of the preset brand can be sent to The user, to obtain the user's recognition result of the target item, if the user's recognition result of the target item is an item that meets the preset condition, then it can be determined that the target item's recognition result is an item that meets the preset condition, if the user's If the recognition result of the target item is that the target item does not meet the preset condition, then it may be determined that the recognition result of the target item is an item that does not meet the preset condition.

In this implementation, the recognition result of the target item by the user is obtained in response to determining that the matching result is a matching failure; The recognition result of the target item is determined based on the final manual recognition result, which effectively improves the accuracy of the recognition result.

In some optional manners, determining whether the brand of the target item is a preset brand based on image information includes: determining whether the brand of the target item is a preset brand based on a style of the target item included in the image information.

In this implementation, the execution subject can identify whether the brand of the target item is the default brand according to the style of the target item included in the identified image information, and if it is determined to be the default brand, combine the authorized brand information of the target item with the The brand information of the preset brand is matched to obtain the matching result.

Here, the style information is used to indicate the style of the item, including but not limited to: outline frame, size, pattern, color, and material.

In this implementation, based on the style of the target item included in the image information, it is determined whether the brand of the target item is the default brand, and in response to determining that the brand of the target item is the default brand, the authorized brand information of the target item is compared with the default brand. Brand information is matched, and based on the matching results, it is identified whether the target item is an infringing item, which effectively improves the efficiency of identifying whether the target item is an infringing item.

Continue to refer to FIG. 3 , which is a schematic diagram of an application scenario of the identification method according to this embodiment.

In the application scenario in Figure 3, the execution subject 301 actively captures the image information of the target item to be identified on an e-commerce platform, such as a water cup, for example, the image 302 containing the water cup, and authorized brand information, such as N brand Authorized store (determined according to the qualification information of the store corresponding to the target item) 303. Further, the executive body can determine whether the brand of the target item is the default brand based on the image information of the target item. If it is determined that the brand of the target item is the default brand, For example, the M brand 304 matches the authorized brand information of the target item: N brand authorized store 303 with the brand information of the preset brand, such as the M brand store 305, and determines the identification of the target item based on the matching result 306. As a result, for example, if the matching result is no match, it is determined that the target item is a non-compliant item.

In the identification method of the present application, by acquiring image information and authorized brand information of the target item; determining whether the brand of the target item is a preset brand based on the image information, in response to determining that the brand of the target item is a preset brand, the authorized brand of the target item The information is matched with the brand information of the preset brand; based on the matching result, the recognition result of the target item is determined, which improves the accuracy and efficiency of recognition.

Further referring to FIG. 4 , it shows a flow 400 of another embodiment of the identification method. In this embodiment, the process 400 of the identification method of this embodiment may include the following steps:

Step 401, acquire the image information and authorized brand information of the target item.

In this embodiment, for implementation details and technical effects of step 401, reference may be made to the description of step 201, and details are not repeated here.

Step 402, based on the logo of the target item included in the image information, determine whether the brand of the target item is a preset brand.

In this embodiment, the execution subject may determine whether the brand of the target item is a preset brand according to the logo of the target item included in the recognized image information, that is, the brand logo.

Here, it should be pointed out that if it is detected that the image information includes only one logo, the logo can be directly determined as the logo of the target item; if it is detected that the image information includes multiple logos, it can be further combined with the class The item information determines the logo of the target item.

In some optional ways, the logo of the target item in the image information is determined by: identifying the logo and the position of the logo included in the image information; based on the category information of the target item, determining the target item in the image information The position of the target item is determined based on the sign, the position of the sign and the position of the target item.

In this implementation, if the image information of the target item includes at least two signs, the execution subject can first identify all the signs and the positions of all signs included in the image information, and then identify Find the position of the target item in the image information, match all the identified marks and the positions of all marks with the position of the target item, and determine the mark corresponding to the position of the mark overlapping with the position of the target item as the mark of the target item.

Here, the executive body can adopt the image recognition algorithm in the existing technology or future development technology, for example, YOLOv4 (You Only Look Once version 4, only look at once recognition version 4) algorithm, faster-rcnn algorithm, etc., to identify the image of the target item All signs and the location of all signs in the information.

Among them, the executive body can use the text recognition algorithm in the existing technology or the future development technology, for example, the TextCNN algorithm, the TextRNN algorithm, etc., to determine the category of the target item indicated by the category information, and then according to the category of the target item. Image recognition algorithms, such as YOLOv4 algorithm, faster-rcnn algorithm, etc., identify the location of the target item.

Specifically, the target item is shorts, and the image indicated by the image information of the target item includes shorts and sneakers, and both the shorts and the sneakers are marked with brand logos, the execution subject can first use the recognition algorithm to identify the All brand logos, that is, the logo of shorts and sports shoes, and the location of the logo, that is, the location area determined according to the detection frame of the logo, and according to the category information of the target item, that is, pants, identify the target item in the image information Position, that is, the position area determined according to the detection frame of the shorts, and then match the position of the shorts with the position of the brand logo of the shorts and the position of the brand logo of the sports shoes to determine the brand logo of the target item, that is, the shorts.

This implementation method recognizes the sign and the position of the sign included in the image information; determines the position of the target item in the image information based on the category information of the target item; determines the position of the target item based on the sign, the position of the sign and the position of the target item The mark of the target item avoids identifying the background item included in the image information as an infringing item, and further improves the accuracy of identifying the infringing item.

In some optional manners, determining whether the brand of the target item is a preset brand based on the logo of the target item included in the image information includes: identifying and selecting an image of the logo of the target item included in the image information; Obtain the target vector corresponding to the image of the logo; determine whether the brand of the target item is a preset brand based on whether there is a similar vector to the target vector in the preset brand logo vector library.

In this implementation, the executive body can use the image recognition algorithm in the existing technology or the future development technology, for example, YOLOv4 algorithm, faster-rcnn algorithm, etc., to identify and frame the logo of the target item included in the image information According to the preset conditions, the image of the logo is converted into a target vector corresponding to the image, that is, the target image representation vector. In the preset brand logo vector library, it is determined whether there is a similarity with the target vector that meets the preset conditions. A vector is a similar vector, for example, a vector whose similarity is greater than or equal to a preset threshold. If it is determined to exist, it is determined that the brand of the target item is a preset brand corresponding to the similarity vector. If it is determined that it does not exist, it is determined that the brand of the target item is not the default brand.

Here, logo vectors corresponding to preset brands are stored in the brand logo vector library. The brand logo vector library can be constructed based on image representation vectors generated by brand logos selected from public datasets and private datasets. Wherein, the brand logo can be a newly registered or registered brand logo.

This implementation method recognizes and selects the image of the logo of the target item included in the image information; obtains the target vector corresponding to the image of the logo; determines whether there is a similar vector to the target vector in the preset brand logo vector library. Whether the brand of the target item is a preset brand, in response to determining that there is a similar vector of the target vector in the preset brand logo vector library, determine that the brand of the target item is a preset brand corresponding to the similar vector, which improves the recognition of the newly registered brand efficiency.

Step 403: In response to determining that the brand of the target item is a preset brand, match the authorized brand information of the target item with the brand information of the preset brand.

In this embodiment, for implementation details and technical effects of step 403, reference may be made to the description of step 203, which will not be repeated here.

Step 404, based on the matching result, determine the recognition result of the target item.

In this embodiment, for implementation details and technical effects of step 404, reference may be made to the description of step 204, which will not be repeated here.

In the above-mentioned embodiment of the present application, compared with the embodiment corresponding to FIG. 2 , the process 400 of the identification method in this embodiment embodies the determination of whether the brand of the target item is preset based on the logo of the target item included in the image information. brand, in response to determining that the brand of the target item is a preset brand, match the authorized brand information of the target item with the brand information of the preset brand, and identify whether the target item is an infringing item based on the matching result, effectively improving the identification of the target item. Whether it is the efficiency and effectiveness of the infringing article.

Continuing to refer to FIG. 5 , a flow 500 of an embodiment of the processing method according to the present application is shown. The processing method includes the following steps:

Step 501, acquiring the identification result of the item to be processed.

In this embodiment, the execution subject can obtain the identification result of the item to be processed at the local or remote device, wherein the identification result of the item to be processed is the identification result obtained by using the method described in the embodiment corresponding to Figure 2, which is not described here Let me repeat.

Step 502, based on the identification result, perform a processing operation on the item to be processed.

In this embodiment, after the execution subject obtains the identification result of the item to be processed, it can perform a processing operation on the item to be processed according to the identification result.

Here, the processing operations may include reward operations, penalty operations, and the like.

Among them, the reward operation may include adding points to the item to be processed, pre-displaying the page, etc.; the punishment operation may include removing the item to be processed, restricting permissions, deducting points, issuing a breach of contract, etc.

Specifically, if the identification result of the item to be processed is an item that meets a preset condition, for example, an item that is not in compliance, a penalty operation may be performed on the item to be processed.

In some optional manners, based on the recognition result, performing a processing operation on the item to be processed includes: in response to determining that the recognition result satisfies a preset condition, performing multimodal fusion based on the image information and category information of the item to be processed, to obtain A fine-grained vector representation of the item to be processed; storing the fine-grained vector representation to the intercepted item database. In this implementation, after the execution subject obtains the identification result of the item to be processed, if the identification result satisfies the preset condition, that is, the item to be processed is an item that meets the preset condition, for example, a non-compliant item, an infringing item, etc., then Multi-modal fusion can be performed according to the image information and category information of the target item, that is, the image information and category information of the target item are fused to obtain the fine-grained vector representation of the target item, and the fine-grained vector representation is stored in the intercepted item database .

Here, the category information of the target item may be determined according to the title information corresponding to the image of the target item.

Among them, multimodal fusion is used to indicate the combination of two or more modalities in various forms, usually referring to the processing of three modalities of image, text, and voice. Multimodal fusion can obtain rich features information.

The executive body can adopt algorithms in existing technologies or future development technologies, such as algorithms based on Bayesian decision theory, algorithms based on sparse representation theory, algorithms based on deep learning theory, etc., and then multi-modal fusion. Not limited.

Specifically, based on the TextCNN (Convolutional Naural Networks for Sentence Classification, convolutional neural network-based text classification model) model and EfficientNet model, as well as the multimodal fusion of Attention, the execution subject can perform image information and category information of the target item. Multi-modal fusion, obtain the fine-grained vector representation of the target item, and store the fine-grained vector representation in the intercepted item database.

Wherein, the intercepted item database is applied to the release and interception process of items meeting preset conditions.

In this implementation, in response to determining that the recognition result meets the preset conditions, based on the image information and category information of the item to be processed, multi-modal fusion is performed to obtain a fine-grained vector representation of the item to be processed; the fine-grained vector representation is stored in the interception The item database helps to strengthen the vector representation of the target item in the image information, weaken the interference of non-target item factors in the image information, and then improve the effect of intercepting the release of items in the database that meet the preset conditions.

With further reference to Figure 6, as an implementation of the methods shown in the above figures, the present application provides an embodiment of an identification device, which corresponds to the method embodiment shown in Figure 2, and the device can specifically be applied to in various electronic devices.

As shown in FIG. 6 , the identification device 600 of this embodiment includes: an acquisition module 601 , a determination module 602 , a matching module 603 and an identification module 604 .

Wherein, the obtaining module 601 may be configured to obtain image information and authorized brand information of the target item.

The determining module 602 may be configured to determine whether the brand of the target item is a preset brand based on the image information.

The matching module 603 may be configured to match the authorized brand information of the target item with the brand information of the preset brand in response to determining that the brand of the target item is a preset brand.

The recognition module 604 is configured to determine the recognition result of the target item based on the matching result.

In some optional manners of this embodiment, the matching module is further configured to: determine whether the brand of the target item is a preset brand based on the logo of the target item included in the image information.

In some optional ways of this embodiment, the logo of the target item is determined by: identifying the logo and the position of the logo included in the image information; determining the target in the image information based on the category information of the target item The location of the object; determining the symbol of the target object based on the symbol, the location of the symbol, and the location of the target object.

In some optional manners of this embodiment, the determination module is further configured to: identify and frame the image of the logo of the target item included in the image information; acquire the target vector corresponding to the image of the logo; Whether there is a similar vector of the target vector in the brand logo vector library of the target item, and determine whether the brand of the target item is a preset brand.

In some optional manners of this embodiment, the determination module is further configured to: determine whether the brand of the target item is a preset brand based on the style of the target item included in the image information.

In some optional forms of this embodiment, the identification module is further configured to: in response to determining that the matching result is a matching failure, obtain the user's identification result of the target item; based on the user's identification result of the target item, determine the target item's recognition result.

Further referring to FIG. 7, as an implementation of the methods shown in the above figures, the present application provides an embodiment of a processing device, which corresponds to the method embodiment shown in FIG. in various electronic devices.

As shown in FIG. 7 , the processing device 700 of this embodiment includes: a processing module 701 and an execution module 702 .

The processing module 701 may be configured to acquire the identification result of the item to be managed.

The execution module 702 may be configured to perform governance operations on the items to be managed based on the identification result.

In some optional ways of this embodiment, the execution module is further configured to: in response to determining that the recognition result satisfies the preset condition, based on the image information and category information of the item to be managed, perform multimodal fusion to obtain the Fine-grained vector representations of items; store the fine-grained vector representations to the governance item database.

Further referring to FIG. 8 , as an implementation of the methods shown in the above figures, the present application provides an embodiment of a smart inspection system. As shown in FIG. 8 , the smart inspection system 800 of this embodiment includes: a service module 801 , a detection module 802 , a cloud storage module 803 , an audit system backend module 804 and an external processing module 805 .

The service module 801 is used to obtain item data of the target item from the database by using a big data tool, wherein the item data includes image information and authorized brand information of the target item. In addition, the item data may also include category information of the target item, store id and other information.

After the service module 801 acquires the item data of the target item and performs preliminary processing, it can send the image information of the target item to the detection module to detect whether the brand of the target item is a preset brand and obtain the detection result, and respond to determine the target item The brand is a preset brand, and based on the brand information of the preset brand and the authorized brand information, the recognition result of the target item is generated, and the recognition result is stored in the cloud storage module.

The detection module 802 is used to detect whether the brand of the target item is a preset brand based on the image information of the target item, and send the detection result to the service module.

The cloud storage module 803 is used for storing the recognition result of the target item.

The audit system back-end module 804 is used to obtain data readiness status information, obtain the identification result of the target item from the cloud storage module, and regularly push the identification result to the external processing module.

The external processing module 805 is configured to perform a preset processing operation on the target item according to the recognition result of the target item, and generate a final confirmation result of the target item (such as a preset processing operation during or before or after performing a preset processing operation). Assuming that no appeal or reconsideration request information is received within the time period, the final confirmation result of the target item can be generated).

In some optional forms of this embodiment, the system further includes: an external image search module 806 , and the detection module 802 includes: a logo detection unit 8021 and a style detection unit 8022 .

The external image search module 806 is used to receive the request sent by the detection module, obtain a list of similar items based on the request, and send the list to the detection module, requesting the style of the target item.

The detection module 802 includes: a logo detection unit 8021 and a style detection unit 8022, wherein the logo detection unit is used to determine the logo of the target article according to the image information of the target article, and determine whether the brand of the target article is a predetermined brand according to the logo of the target article. Set the brand; the style detection unit is used to determine the style of the target item according to the image information of the target item, and send a request to the external image search module to obtain a list of similar items, and based on the list, determine whether the brand of the target item is Default branding.

In some optional manners of this embodiment, the sign detection unit 8021 includes: a general sign detection subunit and a special sign detection subunit.

The general sign detection subunit is used to determine the sign of the target item based on the sign included in the image information of the target item, the position of the sign and the category information of the target item, and determine whether the brand of the target item is the intended brand according to the sign of the target item. Set the brand; the special mark detection subunit is used to determine whether the brand of the target item is preset based on the target vector corresponding to the image of the mark of the target item included in the image information of the target item and the preset brand logo vector library brand.

In some optional manners of this embodiment, the system further includes: an audit front-end page module 807 .

The review front-end page module 807 is used to receive the input item data of the item to be identified, and store the item data into the database.

In some optional ways of this embodiment, the review front-end page module 807 is further configured to visualize the item data of the item to be identified on the review front-end interface for manual review.

In some optional ways of this embodiment, the system further includes: an image vector embedding module 808 and a vector retrieval service module 809 .

The image vector embedding module 808 is configured to obtain the final confirmation result of the target item from an external processing module, the final confirmation result is used to indicate whether the target item is a confirmed infringing item, and in response to determining that the final identification result is a confirmed infringing item, based on the target item The image information generates a picture vector, and stores the picture vector of the target item to the intercepted item database corresponding to the vector retrieval service module;

The vector retrieval service module 809 is configured to search the intercepted item database based on the image vector of the item to be identified, so as to determine whether there is an infringing item similar to the item to be identified.

According to the embodiments of the present application, the present application also provides an electronic device and a readable storage medium.

As shown in FIG. 9 , it is a block diagram of an electronic device according to the identification method of the embodiment of the present application.

900 is a block diagram of an electronic device according to the identification method of the embodiment of the present application. Electronic device is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are by way of example only, and are not intended to limit implementations of the applications described and/or claimed herein.

As shown in FIG. 9, the electronic device includes: one or more processors 901, a memory 902, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are interconnected using different buses and can be mounted on a common motherboard or otherwise as desired. The processor may process instructions executed within the electronic device, including instructions stored in or on the memory, to display graphical information of a GUI on an external input/output device such as a display device coupled to an interface. In other implementations, multiple processors and/or multiple buses may be used with multiple memories and multiple memories, if desired. Likewise, multiple electronic devices may be connected, with each device providing some of the necessary operations (eg, as a server array, a set of blade servers, or a multi-processor system). In FIG. 9, a processor 901 is taken as an example.

The memory 902 is a non-transitory computer-readable storage medium provided in this application. Wherein, the memory stores instructions executable by at least one processor, so that the at least one processor executes the identification method provided in this application. The non-transitory computer-readable storage medium of the present application stores computer instructions, and the computer instructions are used to cause a computer to execute the identification method provided in the present application.

The memory 902, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs and modules, such as the program instructions/modules corresponding to the identification method in the embodiment of the present application (for example, the accompanying drawing 6 shows the acquiring module 601, determining module 602, matching module 603 and identifying module 604). The processor 901 runs non-transitory software programs, instructions and modules stored in the memory 902 to execute various functional applications and identification of the server, that is, to implement the identification method in the above method embodiments.

The memory 902 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application program required by at least one function; the data storage area may store data created by using the identified electronic device, and the like. In addition, the memory 902 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory 902 may optionally include a memory that is remotely located relative to the processor 901, and these remote memories may be connected to the identified electronic device through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The electronic device of the identification method may further include: an input device 903 and an output device 904 . The processor 901, the memory 902, the input device 903, and the output device 904 may be connected through a bus or in other ways, and connection through a bus is taken as an example in FIG. 9 .

The input device 903 can receive input digital or character information, such as a touch screen, a keypad, a mouse, a trackpad, a touch pad, a pointing stick, one or more mouse buttons, a trackball, a joystick and other input devices. The output device 604 may include a display device, an auxiliary lighting device (eg, LED), a tactile feedback device (eg, a vibration motor), and the like. The display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some implementations, the display device may be a touch screen.

Various implementations of the systems and techniques described herein can be implemented in digital electronic circuitry, integrated circuit systems, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpreted on a programmable system including at least one programmable processor, the programmable processor Can be special-purpose or general-purpose programmable processor, can receive data and instruction from storage system, at least one input device, and at least one output device, and transmit data and instruction to this storage system, this at least one input device, and this at least one output device an output device.

These computing programs (also referred to as programs, software, software applications, or codes) include machine instructions for a programmable processor and may be implemented using high-level procedural and/or object-oriented programming languages, and/or assembly/machine language calculation program. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or means for providing machine instructions and/or data to a programmable processor ( For example, magnetic disks, optical disks, memories, programmable logic devices (PLDs), including machine-readable media that receive machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.

To provide for interaction with the 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 identifying the user; and a keyboard and pointing device (eg, a mouse or trackball) through which a user can provide input to the computer. Other kinds 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 can be in any form (including Acoustic input, speech input or, tactile input) to receive input from the user.

The systems and techniques described herein can 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., as a a user computer having a graphical user interface or web browser through which a user can interact with embodiments of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system can be interconnected by any form or medium of digital data communication, eg, a communication network. Examples of communication networks include: Local Area Network (LAN), Wide Area Network (WAN) and the Internet.

A computer system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other.

According to the technical solutions of the embodiments of the present application, the accuracy and efficiency of recognition are effectively improved.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, the steps described in the present application may be executed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution disclosed in the present application can be achieved, no limitation is imposed herein.

The above specific implementation methods are not intended to limit the protection scope of the present application. It should be apparent to those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made depending on design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application shall be included within the protection scope of this application.

Claims

An identification method, the method comprising:

Obtain the image information and authorized brand information of the target item;

Based on the image information, determine whether the brand of the target item is a preset brand;

In response to determining that the brand of the target item is a preset brand, matching the authorized brand information of the target item with the brand information of the preset brand;

Based on the matching result, the recognition result of the target item is determined.
The method according to claim 1, wherein, based on the image information, determining whether the brand of the target item is a preset brand comprises:

Based on the logo of the target item included in the image information, it is determined whether the brand of the target item is a preset brand.
The method according to claim 2, wherein the mark of the target item is determined by:

identifying the logo and the location of the logo included in the image information;

determining the position of the target item in the image information based on the category information of the target item;

An identifier of the target item is determined based on the marker, the position of the marker, and the position of the target item.
The method according to claim 2, wherein, based on the logo of the target item included in the image information, determining whether the brand of the target item is a preset brand comprises:

identifying and frame-selecting the image of the logo of the target item included in the image information;

obtaining an object vector corresponding to the image of the sign;

Based on whether there is a similar vector to the target vector in the preset brand logo vector library, it is determined whether the brand of the target item is a preset brand.
The method according to claim 1, wherein, based on the image information, determining whether the brand of the target item is a preset brand comprises:

Based on the style of the target item included in the image information, it is determined whether the brand of the target item is a preset brand.
The method according to any one of claims 1-5, wherein the determining the identification result of the target item based on the matching result comprises:

In response to determining that the matching result is a matching failure, acquiring a user's recognition result of the target item;

Based on the recognition result of the target item by the user, the recognition result of the target item is determined.
The method according to any one of claims 1-5, wherein the determining the identification result of the target item based on the matching result comprises:

In response to determining that the matching result is a matching failure, it is determined that the target item is an infringing item.
A treatment method, said method comprising:

Acquiring the identification result of the item to be processed, the identification result is the identification result obtained by identifying the item to be processed by using the identification method described in any one of claims 1-7;

Based on the identification result, a processing operation is performed on the item to be processed.
The method according to claim 8, wherein said performing a processing operation on said item to be processed based on said identification result comprises:

In response to determining that the recognition result satisfies a preset condition, multimodal fusion is performed based on the image information and category information of the item to be processed to obtain a fine-grained vector representation of the item to be processed;

The fine-grained vector representation is stored in an intercepted item database, and the intercepted item database is applied to the release interception process of items meeting preset conditions.
An identification device, said device comprising:

An acquisition module configured to acquire image information and authorized brand information of the target item;

A determining module configured to determine whether the brand of the target item is a preset brand based on the image information;

A matching module configured to match the authorized brand information of the target item with the brand information of the preset brand in response to determining that the brand of the target item is a preset brand;

The recognition module is configured to determine the recognition result of the target item based on the matching result.
The apparatus of claim 10, wherein the determining module is further configured to:

Based on the logo of the target item included in the image information, it is determined whether the brand of the target item is a preset brand.
The apparatus of claim 11, wherein the identity of the target item is determined by:

identifying the logo and the location of the logo included in the image information;

determining the position of the target item in the image information based on the category information of the target item;

An identifier of the target item is determined based on the marker, the position of the marker, and the position of the target item.
The apparatus of claim 11, wherein the determining module is further configured to:

identifying and frame-selecting the image of the logo of the target item included in the image information;

obtaining an object vector corresponding to the image of the sign;

Based on whether there is a similar vector to the target vector in the preset brand logo vector library, it is determined whether the brand of the target item is a preset brand.
The apparatus of claim 10, wherein the determining module is further configured to:

Based on the style of the target item included in the image information, it is determined whether the brand of the target item is a preset brand.
The device according to any one of claims 10-14, wherein the identification module is further configured to:

In response to determining that the matching result is a matching failure, acquiring a user's recognition result of the target item;

Based on the recognition result of the target item by the user, the recognition result of the target item is determined.
The device according to any one of claims 10-14, wherein the identification module is further configured to:

In response to determining that the matching result is a matching failure, it is determined that the target item is an infringing item.
A processing device, said device comprising:

The processing module is configured to obtain the identification result of the item to be processed, the identification result is the identification result obtained by identifying the item to be processed by using the identification method according to any one of claims 1-7;

An execution module configured to execute a processing operation on the item to be processed based on the recognition result.
The apparatus of claim 17, wherein the execution module is further configured to:

In response to determining that the recognition result satisfies a preset condition, multimodal fusion is performed based on the image information and category information of the item to be processed to obtain a fine-grained vector representation of the item to be processed;

The fine-grained vector representation is stored in an intercepted item database, and the intercepted item database is applied to the release interception process of items meeting preset conditions.
A kind of intelligent detection system, described system comprises:

The service module is used to apply the big data tool to obtain the item data of the target item from the database, the item data includes the image information of the target item and the authorized brand information, and sends the image information of the target item to the detection module to detect the target item Whether the brand of the target item is a preset brand and obtain the detection result, and in response to determining that the brand of the target item is a preset brand, based on the brand information of the preset brand and the authorized brand information, an identification result of the target item is generated, and the identification result Store to the cloud storage module;

A detection module, configured to detect whether the brand of the target item is a preset brand based on the image information of the target item, and send the detection result to the service module;

The cloud storage module is used to store the recognition result of the target item;

The audit system back-end module is used to obtain the identification result of the target item from the cloud storage module, and regularly and uniformly push the identification result to the external governance module;

The external processing module is configured to perform preset processing operations on the target item according to the identification result of the target item, and generate a final confirmation result of the target item.
The system according to claim 19, further comprising: an external image search module, configured to receive a request sent by the detection module, and obtain a list of similar items based on the request, and send the list to the detection module , the request includes the style of the target item;

The detection module includes: a mark detection unit and a style detection unit,

The logo detection unit is configured to determine the logo of the target article according to the image information of the target article, and determine whether the brand of the target article is a preset brand according to the logo of the target article;

The style detection unit is configured to determine the style of the target item according to the image information of the target item, and send a request to the external image search module to obtain a list of similar items, and determine the target based on the list Whether the item's brand is the default brand.
The system of claim 20, wherein the marker detection unit comprises:

The general sign detection subunit is used to determine the sign of the target item based on the sign included in the image information of the target item, the position of the sign and the category information of the target item, and determine the brand of the target item according to the sign of the target item Whether it is the default brand;

The dedicated sign detection subunit is used to determine whether the brand of the target item is a preset brand based on the target vector corresponding to the image of the target item included in the image information of the target item and the preset brand logo vector library.
The system of claim 19, further comprising:

The audit front-end page module is used to receive the input item data of the item to be identified, and store the item data into the database.
The system according to claim 22, wherein the review front-end page module is further used to visualize the item data of the item to be identified on the review front-end interface for manual review.
The system according to claim 19, said system further comprising: an image vector embedding module and a vector retrieval service module, wherein,

The image vector embedding module is used to obtain the final confirmation result of the target item from the external processing module, the final confirmation result is used to indicate whether the target item is a confirmed infringing item, and in response to determining that the final confirmation result is a confirmed infringing item, based on the target item image information to generate a picture vector, and store the picture vector of the target item in the intercepted item database corresponding to the vector retrieval service module;

The vector retrieval service module is used to search the intercepted item database based on the image vector of the item to be identified, so as to determine whether there is an infringing item similar to the item to be identified.
An electronic device, characterized in that it comprises:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

The memory stores information executable by the at least one processor, so that the at least one processor can execute the method according to any one of claims 1-9.
A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to make the computer execute the method according to any one of claims 1-9.