CN112613891A - Shop registration information verification method, device and equipment - Google Patents

Shop registration information verification method, device and equipment Download PDF

Info

Publication number
CN112613891A
CN112613891A CN202011572760.1A CN202011572760A CN112613891A CN 112613891 A CN112613891 A CN 112613891A CN 202011572760 A CN202011572760 A CN 202011572760A CN 112613891 A CN112613891 A CN 112613891A
Authority
CN
China
Prior art keywords
shop
verified
information
analysis result
store
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202011572760.1A
Other languages
Chinese (zh)
Other versions
CN112613891B (en
Inventor
章云帆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alipay Hangzhou Information Technology Co Ltd
Original Assignee
Alipay Hangzhou Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alipay Hangzhou Information Technology Co Ltd filed Critical Alipay Hangzhou Information Technology Co Ltd
Priority to CN202011572760.1A priority Critical patent/CN112613891B/en
Publication of CN112613891A publication Critical patent/CN112613891A/en
Application granted granted Critical
Publication of CN112613891B publication Critical patent/CN112613891B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/018Certifying business or products
    • G06Q30/0185Product, service or business identity fraud
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • G06F18/241Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00Machine learning
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/22Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/40Scenes; Scene-specific elements in video content
    • G06V20/41Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/60Type of objects
    • G06V20/62Text, e.g. of license plates, overlay texts or captions on TV images

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Business, Economics & Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Multimedia (AREA)
  • Software Systems (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computational Linguistics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Medical Informatics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Finance (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiment of the specification discloses a method, a device and equipment for verifying store registration information. The scheme comprises the following steps: extracting text information and image information related to-be-verified registration information of a to-be-verified shop from shop video data of the to-be-verified shop; processing the text information and the image information by utilizing a multi-modal model to obtain a registration information analysis result for the shop to be verified; and performing consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result of the to-be-verified registration information of the to-be-verified shop.

Description

Shop registration information verification method, device and equipment
Technical Field
The application relates to the technical field of internet, in particular to a method, a device and equipment for verifying shop registration information.
Background
Currently, merchants may register individual physical stores at a commerce platform to facilitate commerce at the commerce platform. The commerce platform needs to verify the authenticity of the store registration information submitted by the merchant to avoid adverse effects on the consumer due to the merchant using false information.
Therefore, the problem that needs to be solved urgently is how to provide a verification method for store registration information with better accuracy and effectiveness.
Disclosure of Invention
The embodiment of the specification provides a method, a device and equipment for verifying store registration information, and aims to improve the accuracy and the validity of an authenticity verification result aiming at the store registration information to be verified.
In order to solve the above technical problem, the embodiments of the present specification are implemented as follows:
the shop registration information verification method provided by the embodiment of the specification comprises the following steps:
obtaining shop video data of a shop to be verified; the shop to be verified is a shop with registration information to be verified;
extracting text information related to the to-be-verified registration information from the shop video data;
extracting image information related to the to-be-verified registration information from the shop video data;
processing the text information and the image information by using a multi-modal model to obtain a registration information analysis result for the shop to be verified;
and performing consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result aiming at the to-be-verified registration information.
A store registration information verification apparatus provided in an embodiment of the present specification includes:
the system comprises an acquisition module, a verification module and a verification module, wherein the acquisition module is used for acquiring shop video data of shops to be verified; the shop to be verified is a shop with registration information to be verified;
the first extraction module is used for extracting text information related to the to-be-verified registration information from the shop video data;
the second extraction module is used for extracting image information related to the to-be-verified registration information from the shop video data;
the analysis module is used for processing the text information and the image information by utilizing a multi-modal model to obtain a registration information analysis result aiming at the shop to be verified;
and the verification module is used for performing consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result aiming at the to-be-verified registration information.
A store registration information verification apparatus provided in an embodiment of the present specification includes:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
obtaining shop video data of a shop to be verified; the shop to be verified is a shop with registration information to be verified;
extracting text information related to the to-be-verified registration information from the shop video data;
extracting image information related to the to-be-verified registration information from the shop video data;
processing the text information and the image information by using a multi-modal model to obtain a registration information analysis result for the shop to be verified;
and performing consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result aiming at the to-be-verified registration information.
At least one embodiment provided in the present specification can achieve the following advantageous effects:
by extracting the text information and the image information related to the to-be-verified registration information of the to-be-verified shop from the continuous dynamic shop video data, compared with the conventional shop related information extracted from the shop certificate image and the shop door head image, the information amount of the extracted shop related information can be increased, and the credibility of the extracted shop related information can be improved by increasing the counterfeiting cost of the shop related information. Subsequently, the text information and the image information are processed by utilizing a multi-mode model, so that the image-text information can be subjected to fusion analysis, and the accuracy and the effectiveness of the generated registration information analysis result for the shop to be verified can be improved; and then the accuracy and the effectiveness of the verification result generated based on the registration information analysis result and aiming at the registration information to be verified can be improved.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort.
FIG. 1 is a schematic view of a swim lane flow of a method for verifying store registration information according to an embodiment of the present disclosure;
fig. 2 is a schematic flow chart of a method for verifying store registration information according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a store registration information verification apparatus corresponding to fig. 2 provided in an embodiment of the present specification;
fig. 4 is a schematic structural diagram of a store registration information verification apparatus corresponding to fig. 2 provided in an embodiment of the present specification.
Detailed Description
To make the objects, technical solutions and advantages of one or more embodiments of the present disclosure more apparent, the technical solutions of one or more embodiments of the present disclosure will be described in detail and completely with reference to the specific embodiments of the present disclosure and the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present specification, and not all embodiments. All other embodiments that can be derived by a person skilled in the art from the embodiments given herein without making any creative effort fall within the scope of protection of one or more embodiments of the present specification.
The technical solutions provided by the embodiments of the present description are described in detail below with reference to the accompanying drawings.
In the prior art, a business platform (i.e., an electronic business platform) generally refers to a platform that provides an online transaction negotiation environment for an enterprise or an individual. The business platform can provide a virtual network space for business activities established on the Internet and a management environment for ensuring smooth operation of the business; the system is an important place for coordinating and integrating information flow, cargo flow and fund flow in order, relevance and high-efficiency flow. Enterprises and merchants can make full use of shared resources such as network infrastructure, payment platform, security platform, management platform and the like provided by the electronic commerce platform to effectively develop own commercial activities at low cost.
The merchant can register the personal off-line physical store at the commerce platform, and submit the store business license image or the store head photo and the like as the reference information of the store registration information, if the store registration information submitted by the merchant passes the confidence verification at the commerce platform, the commerce platform can open the electronic business card for the off-line physical store, so that the merchant can carry out on-line operation activities through the commerce platform. Because the image information submitted by the merchant is low in counterfeiting cost and the information such as the shop address and the industry to which the shop belongs cannot be accurately verified based on the image submitted by the merchant, the accuracy and the effectiveness of the verification result generated aiming at the registration information of the shop to be verified are influenced.
In order to solve the defects in the prior art, the scheme provides the following embodiments:
fig. 1 is a schematic view of a swim lane flow of a store registration information verification method provided in an embodiment of the present specification. As shown in fig. 1, the store registration information verification process involves a merchant device and a verification device. The merchant device and the verification device may be the same device or different devices.
Assuming that the merchant device and the verification device are different devices, in the store video acquisition stage, the merchant may use the merchant device to acquire store video data for the store to be verified, and the merchant may also send the store video data acquired by the merchant device to the verification device.
In the stage of verifying the store registration information, the verification device can acquire the store video data of the store to be verified, which is acquired by a merchant through merchant equipment, and the information quantity and the authenticity of the text information and the image information related to the registration information to be verified, which are extracted from the store video data, can be improved because the information quantity of the store related information contained in the continuous dynamic store video data is larger and the authenticity is better. By utilizing the multi-mode model to process the text information and the image information, the image-text information can be subjected to fusion analysis, and the accuracy and the effectiveness of the generated registration information analysis result for the shop to be verified can be improved; and after consistency verification is carried out on the registration information analysis result and the to-be-verified registration information, the accuracy and the effectiveness of the verification result of the to-be-verified registration information of the to-be-verified shop can be improved.
Fig. 2 is a schematic flow chart of a method for verifying store registration information according to an embodiment of the present disclosure. From a program perspective, the executing subject of the flow may be at least one of a merchant device, a store registration information verification application installed at the merchant device, a server of the store registration information verification application, a device at a commerce platform, or the like. The store registration information verification application may be implemented by an application (e.g., a payment application, a shopping application) provided by the commerce platform for providing a service for the business of the merchant. As shown in fig. 2, the process may include the following steps:
step 202: obtaining shop video data of a shop to be verified; the shop to be verified is a shop with registration information to be verified.
In the embodiment of the specification, a merchant may perform registration operation on a store to be verified at a commerce platform and submit registration information to be verified for the store to be verified, where the registration information to be verified may include store name information, industry information to which the store belongs, store address information, and the like of the store to be verified. If the business platform verifies the registration information to be verified, the merchant can conduct business activities for the shop to be verified at the business platform.
In this embodiment of the present specification, the merchant further needs to collect the store video data of the store to be verified by using the target application, and send the collected store video data to the executive main in the flow in fig. 2, so that the executive main in the flow in fig. 2 verifies the registration information to be verified of the store to be verified based on the obtained store video data. The target application program may be a client of a payment application or a shopping application provided by the commerce platform.
Specifically, in the process that a merchant collects the shop video data by using a target application program, the target application program can prompt the merchant to respectively shoot contents such as a signboard, a shop interior scene, a shop outside street scene, a price list and the like of a shop to be verified in each specified time period in a bullet screen or voice mode and the like, and the merchant can also introduce the shop to be verified in the process of collecting the shop video data in a voice mode; therefore, richer shop related information can be conveniently acquired through the mode of the small video. In practical applications, the whole process of acquiring the video data of the store may be set to tens of seconds to minutes, and after receiving an upload instruction of the merchant for the video data of the store, the target application may send the acquired video data of the store to the execution subject of the process in fig. 2. The reality and the credibility of the store video data acquired in step 202 are improved by making the store video data acquired by the execution subject in the flow in fig. 2 be data that is acquired and uploaded by the merchant in real time by using the target application program.
In practical application, whether illegal behaviors such as counterfeiting and tampering exist in the shop video data can be identified by detecting differences (for example, noise distribution differences, PCA main direction differences, edge pixel point distribution differences and the like) between a foreground region and a background region in the shop video data, and if yes, the shop video data is skipped to the end so as to ensure the accuracy and the validity of a generated verification result.
Or whether illegal behaviors such as screen capture, screen shooting and the like exist in the shop video data can be identified through a classification model built based on a convolutional neural network, in addition, picture watermark detection can be carried out through a target detection algorithm to identify whether illegal behaviors such as the use of illegal video data exist, and if yes, the process is skipped to the end to ensure the accuracy and the validity of the generated verification result.
Step 204: and extracting text information related to the to-be-verified registration information from the shop video data.
In this embodiment of the present specification, an image with a text may be obtained from store video data, and then text information may be extracted from the image based on an OCR (Optical Character Recognition) technique, so as to determine text information related to registration information to be verified. In practical application, because the video acquisition process can be provided with the prompt information aiming at the store information to be acquired in the form of a bullet screen or voice, the text information related to the specified registration information to be verified can be extracted from the specified time period of the store video data according to the display time of each prompt information.
For example, assuming that the presentation time of the prompt information for prompting collection of the tariff information is 1 minute 10 seconds, and the presentation time of the prompt information for prompting collection of the street view outside the shop is 1 minute 30 seconds, it is possible to intercept an image from the shop video data between 1 minute 10 seconds and 1 minute 30 seconds, and extract the tariff text information from the intercepted image. Since the price list may contain text information related to the business information to which the store to be verified belongs, the extracted price list text information may be used as text information related to the business information to which the store to be verified belongs.
Step 206: and extracting image information related to the to-be-verified registration information from the shop video data.
Similarly, image information related to registration information to be authenticated may be extracted from video frames included in the shop video data.
For example, assuming that the display time for prompting to collect the prompting message of the street view outside the shop is 1 minute and 30 seconds, and then no other prompting message exists, the image can be captured from the video data of the shop after 1 minute and 30 seconds to obtain the image of the street view outside the shop. Since information related to the store address information to be verified may be included in the store outside street view image, the extracted store outside street view image may be used as image information related to the store address information to be verified.
Step 208: and processing the text information and the image information by utilizing a multi-modal model to obtain a registration information analysis result aiming at the shop to be verified.
In the embodiments of the present specification, each source or form of information may be referred to as a modality. Such as audio, video, text, images, etc. Since the modality generally has a very broad definition, the data type corresponding to the information of each modality is not particularly limited.
The multi-modal model is a model constructed based on the principle of multi-modal Machine Learning (MMML), and aims to realize the capability of processing and understanding multi-source modal information by a Machine Learning method. In practical application, the multi-modal model can eliminate redundancy among modal information by utilizing complementarity among the multi-modal information to learn more accurate feature representation, and further can apply the feature representation with better accuracy generated by the multi-modal model to the fields of information retrieval, classification task processing, regression task processing and the like so as to improve the accuracy of the obtained information retrieval result, classification task processing result and regression task processing result. In the embodiment of the specification, the accuracy of the registration information analysis result generated by using the multi-modal model is better.
Step 210: and performing consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result aiming at the to-be-verified registration information.
In this embodiment of the present specification, the registration information analysis result obtained in step 208 may be determined as reference information with higher reliability, and if the registration information to be verified is consistent with the registration information analysis result, it may be indicated that the registration information to be verified is authentic and authentic, so that a verification result that the registration information to be verified passes verification may be generated; if the to-be-verified registration information is inconsistent with the registration information analysis result, the confidence level of the to-be-verified registration information submitted by the merchant in advance can be indicated to be poor, and a verification result that the to-be-verified registration information cannot be verified can be generated.
In the scheme in fig. 2, by extracting the text information and the image information related to the to-be-verified registration information of the to-be-verified shop from the continuously dynamic shop video data, compared with the conventional extraction of the shop related information from the shop certificate image and the shop head image, the information amount of the extracted shop related information can be increased, and the reliability of the extracted shop related information can be improved by increasing the counterfeiting cost of the shop related information. Subsequently, the text information and the image information are processed by utilizing a multi-mode model, so that the image-text information can be subjected to fusion analysis, and the accuracy and the effectiveness of the generated registration information analysis result for the shop to be verified can be improved; and then the accuracy and the effectiveness of the verification result generated based on the registration information analysis result and aiming at the registration information to be verified can be improved.
Based on the method of fig. 2, the present specification also provides some specific embodiments of the method, which are described below.
In the embodiment of the present specification, the to-be-verified registration information submitted by the merchant in advance may include industry information (that is, preset industry information) to which the to-be-verified store belongs, so that an industry analysis result for the to-be-verified store needs to be generated based on store video data of the to-be-verified store, so as to perform subsequent verification.
When the preset industry information in the registration information to be verified needs to be verified, the text information extracted in step 204 needs to include a first text related to the business scope of the store. The image information extracted in step 206 includes a first image related to the business scope of the store.
Correspondingly, step 208: processing the text information and the image information by using a multi-modal model to obtain a registration information analysis result for the shop to be verified, which may specifically include:
and processing the first text and the first image by using a first multi-modal model to obtain an industry analysis result for the shop to be verified.
In the present specification embodiment, the first multimodal model may be trained in advance using training samples. Specifically, a first text and a first image collected from a real off-line shop may be used as a training sample, and an industry to which the real off-line shop belongs may be set as an output label corresponding to the training sample. And obtaining the trained first multi-modal model by taking the positive sample as input information of the first multi-modal model and taking the output label corresponding to the positive sample as a reference output result of the first multi-modal model. In practical application, a negative sample can be further set to train the first multi-modal model, which is not described herein again. The trained first multi-modal model can predict the industry information of the shop corresponding to the text and the graphics according to the input text and the graphics information.
In an embodiment of the present specification, the first text may specifically include at least one of a store signboard text and a store price list text. The shop signboard text can be used for 'XX supermarket', 'XX clinic', 'XX chafing dish', and the like, and can reflect that the industry to which the shop to be verified belongs is retail industry, medical industry or catering industry. The store price list text can be "XX ball pen", "XX Yuan blood routine test", "XX Yuan fish-flavor pork, and it can be seen that the store price list text can also reflect the industry of the store to be verified.
The first image may specifically include at least one of an image of an exterior of a store and an image of an interior of the store. The image of the interior of the shop may be, for example, an image of a shelf in a supermarket, an image of a restaurant including a scene of dining for people, an image of a hospital consulting room, or the like. And the image of the external scene of the shop can be an image of the shop front, an image of the street scene around the shop, etc. It can be known that the first image may also reflect industry information to which the store to be authenticated belongs.
In practical applications, since the store video data may further include first audio data related to the store operation range, for example, an introduction voice of a merchant included in the store video data to the store operation range of the store to be authenticated and an industry of the store to be authenticated, a call of the store to be authenticated included in the store video data, a promotion voice, and the like. Therefore, step 208 may specifically further include:
and processing at least two data of the first text, the first image and the first audio data by using a first multi-modal model to obtain an industry analysis result for the shop to be verified.
Correspondingly, the training sample of the first multimodal model may also be the first text, the first image and/or the first audio data collected from the real off-line shop.
In the implementation mode, the industry to which the store to be verified belongs is predicted by combining audio data on the basis of information such as images and texts, so that the content of the extracted store-related information is richer, and the accuracy of the generated industry analysis result for the store to be verified is improved.
In this embodiment, when the industry analysis result obtained in step 208 is for the store to be verified, step 210 may specifically include:
and judging whether the industry analysis result and preset industry information in the to-be-verified registration information belong to the same industry or not to obtain a first judgment result.
And generating a verification result aiming at the preset industry information according to the first judgment result.
The determining whether the industry analysis result and the preset industry information in the registration information to be verified belong to the same industry may specifically include:
and determining a first credibility that the preset industry information in the industry analysis result and the registration information to be verified belongs to the same industry according to preset industry classification information.
And judging whether the first credibility is larger than a first threshold value.
In the embodiment of the present specification, the preset industry classification information may be implemented by using the existing industry classification information, or may be set by itself according to actual requirements, which is not specifically limited. For example, a text containing industry information may be obtained, irrelevant information such as place names, person names, and auxiliary words may be removed through Word segmentation processing, part-of-speech analysis, denoising processing, and the like, while relevant information of the industry may be retained, and an industry dictionary (i.e., preset industry classification information) may be finally obtained through processing such as Word frequency tagging (english: term frequency-inverse document frequency, english abbreviation: TF-IDF), Word2 Vec-based keyword vectorization, and similar clustering based on the Louvain algorithm. After the industry dictionary is obtained, a first credibility that the industry analysis result and preset industry information in the registration information to be verified belong to the same industry can be determined based on the industry dictionary by using a weighted aggregation principle. Or, it may also be determined whether the industry analysis result and the preset industry information in the registration information to be verified belong to a category of industry in the industry dictionary, if yes, the first reliability may be determined as a value greater than a first threshold, and if not, the first reliability may be determined as a value less than or equal to the first threshold.
Subsequently, if the judgment result indicates that the first credibility is greater than the first threshold, it may be determined that the industry analysis result and the preset industry information in the registration information to be verified belong to the same industry, that is, the preset industry information in the registration information to be verified submitted by the merchant is true and credible, so that a verification result passing verification for the preset industry information may be generated. And if the judgment result shows that the first credibility is less than or equal to the first threshold, generating a verification result which does not pass the verification of the preset industry information.
In the embodiment of the present specification, the to-be-verified registration information submitted by the merchant in advance may further include store name information (i.e., preset store name information) of the store to be verified, so that it is necessary to generate a store name analysis result for the store to be verified based on the store video data of the store to be verified, so as to perform subsequent verification.
When verification needs to be performed on preset shop name information in the registration information to be verified, the text information extracted in step 204 may include a second text related to the shop name; and the image information extracted in step 206 may include a second image associated with the store name.
Correspondingly, step 208: processing the text information and the image information by using a multi-modal model to obtain a registration information analysis result for the shop to be verified, which may specifically include:
and processing the second text and the second image by using a second multi-modal model to obtain a shop name analysis result for the shop to be verified.
In the present specification embodiment, the second multimodal model may be trained in advance using training samples. Specifically, the second text and the second image collected from the real off-line shop may be used as a training sample, and the shop name information of the real off-line shop may be set as the output label corresponding to the training sample. And obtaining a trained second multi-modal model by taking the positive sample as input information of the second multi-modal model and taking the output label corresponding to the positive sample as a reference output result of the second multi-modal model. In practical application, a negative sample can be further set to train the second multi-modal model, which is not described herein again. The trained second multi-modal model can predict the shop name information of the shop corresponding to the text and the graphics according to the input text and the graphics information.
In this specification embodiment, the second text may specifically include a shop signboard text. The shop signboard text can be "XX supermarket", "XX clinic", "XX hot pot", and the like, and can be visible, the shop signboard text can reflect the shop name information of the shop to be verified, and the second image can be the shop signboard image.
In practical applications, since the second audio data related to the store name information may also be included in the store video data, for example, an introduction voice of a merchant to the store name of the store to be authenticated included in the store video data, a call of the store to be authenticated included in the store video data, a publicity voice, and the like. Therefore, step 208 may specifically include:
and processing at least two data of the second text, the second image and the second audio data by using a second multi-modal model to obtain a shop name analysis result for the shop to be verified.
Correspondingly, the training sample of the second multimodal model may also be second text, second image and/or second audio data collected from a real off-line shop.
In the implementation mode, the store name to which the store to be verified belongs is predicted by combining the audio data on the basis of the information such as the image and the text, so that the content of the extracted information related to the store name is richer, and the accuracy of the analysis result of the store name of the store to be verified is improved. Of course, in practical applications, the text information extracted from the shop signboard image may be directly used as the shop name analysis result for the shop to be verified, so as to simplify the verification process.
In this embodiment of the present specification, when the result obtained in step 208 is a result of analyzing the store name of the store to be verified, step 210 may specifically include:
and determining a second credibility of the preset shop name information according to the text similarity between the shop name analysis result and the preset shop name information in the to-be-verified registration information.
And judging whether the second credibility is greater than a second threshold value or not to obtain a second judgment result.
And generating a verification result aiming at the preset shop name information according to the second judgment result.
In this embodiment of the present specification, an existing algorithm for calculating text similarity may be used to calculate the text similarity between the shop name analysis result and the preset shop name information in the registration information to be verified, which is not specifically limited.
In practical application, the calculated text similarity can be directly determined as the second credibility of the preset shop name information, and the method is simple and convenient. Or, a corresponding relationship between the second reliability and a preset value section may be determined in advance according to the relevant data of the sample store, so that the second reliability corresponding to the preset value section to which the calculated text similarity belongs is determined as the second reliability of the preset store name information. In general, the maximum value of the preset value interval is in direct proportion to the corresponding second reliability. For example, the preset value intervals may be [0,0.6], (0.6,0.8], (0.8,1], and the corresponding second confidence levels may be 0.4, 0.7, 0.9, and the like, respectively.
In this embodiment of the specification, if the second determination result indicates that the second reliability is greater than a second threshold, it may be determined that the store name analysis result is consistent with the preset store name information, that is, the preset store name information in the to-be-verified registration information submitted by the merchant is authentic and reliable, so that a verification result that the verification of the preset store name information is passed may be generated. Correspondingly, if the second determination result indicates that the second credibility is less than or equal to the second threshold, a verification result that the preset store name information is not verified can be generated.
In the embodiment of the present specification, the to-be-verified registration information submitted by the merchant in advance may further include store address information (that is, preset store address information) of the to-be-verified store, so that it is necessary to generate a store address analysis result for the to-be-verified store based on the store video data of the to-be-verified store, so as to perform subsequent verification.
When verification needs to be performed on preset store address information in the registration information to be verified, the text information extracted in step 204 may include a third text related to the store address; and the image information extracted in step 206 may include a third image associated with the address of the store.
Correspondingly, step 208: processing the text information and the image information by using a multi-modal model to obtain a registration information analysis result for the shop to be verified, which may specifically include:
and processing the third text and the third image by using a third multi-modal model to obtain a shop address analysis result for the shop to be verified.
In the present specification embodiment, the third multimodal model may be trained in advance using training samples. Specifically, the third text and the third image collected from the real off-line shop may be used as a training sample, and the shop address information of the real off-line shop may be set as the output label corresponding to the training sample. And obtaining a trained third multi-modal model by taking the positive sample as input information of the third multi-modal model and taking the output label corresponding to the positive sample as a reference output result of the third multi-modal model. In practical application, a negative sample can be further set to train a third multi-modal model, which is not described herein again. The trained third multi-modal model can predict the shop address information of the shop corresponding to the text and the graphics according to the input text and the graphics information.
In an embodiment of the present specification, the third text may specifically include at least one of a shop doorplate text and a street signboard text; the shop house plate text can be as "XX road XX number", "XX cell XX number" and the like, and the street sign plate text can be as "south XX road north", "east XX lane west" and the like, so that the shop house plate text and the street sign plate text can both reflect the shop address information of the shop to be verified. And the third image may include: at least one of the shop doorplate image of the shop to be verified, the street sign image around the shop to be verified, and the street view image around the shop to be verified is omitted for further description.
In practical applications, since the store video data may further include third audio data related to the store address information, for example, an introduction voice of the merchant to the store address of the store to be authenticated included in the store video data, a selling and publicizing voice of the store to be authenticated included in the store video data, and the like. Therefore, step 208 may specifically include:
and processing at least two data of the third text, the third image and the third audio data by using a third multi-modal model to obtain a shop name analysis result for the shop to be verified.
Correspondingly, the training sample of the third multimodal model may also be third text, third image and/or third audio data collected from a real off-line shop.
In the implementation mode, the store address to which the store to be verified belongs is predicted by combining audio data on the basis of information such as images and texts, so that the content of the relevant information of the extracted store address is richer, and the accuracy of the analysis result of the store address of the store to be verified is improved.
Of course, in practical application, the text extracted from the shop doorplate image of the shop to be verified or the street sign image around the shop to be verified can also be directly used as the shop address analysis result for the shop to be verified; in addition, the position information of the shop where the merchant collects the shop video data of the shop to be verified can be obtained and directly used as the shop address analysis result of the shop to be verified, and the verification process can be simplified.
In this embodiment of the present specification, when the result obtained in step 208 is a result of analyzing the store address of the store to be verified, step 210 may specifically include:
and determining a third credibility of the preset shop address information according to a distance value between the geographical position indicated by the shop address analysis result and the geographical position indicated by the preset shop address information in the to-be-verified registration information.
And judging whether the third reliability is greater than a third threshold value or not to obtain a third judgment result.
And generating a verification result aiming at the preset shop address information according to the third judgment result.
In the embodiment of the present specification, a straight-line distance value between the shop address analysis result and the geographical position indicated by the preset shop address information may be calculated from the map data. Typically, the calculated distance value is inversely proportional to the third confidence level of the preset store address information.
In this embodiment of the specification, if the third determination result indicates that the third reliability is greater than a third threshold, it may be determined that the store address analysis result is consistent with the preset store address information, that is, the preset store address information in the to-be-verified registration information submitted by the merchant is authentic and reliable, so that a verification result that the verification of the preset store address information passes may be generated. Correspondingly, if the third determination result indicates that the third credibility is less than or equal to the third threshold, a verification result that the preset store address information is not verified can be generated.
In the above-described embodiments in the present specification, an implementation is provided in which analysis results for various types of registration information of a store to be authenticated are generated from data extracted from store video data of the store to be authenticated. Since various types of registration information of merchants usually have relevance, consistency verification can be performed on target types of registration information to be verified based on other types of registration information analysis results, so that the accuracy of verification results is improved.
Specifically, when verifying the preset industry information in the registration information to be verified, after determining the first credibility that the industry analysis result and the preset industry information in the registration information to be verified belong to the same industry, the method may further include:
and obtaining a shop name analysis result aiming at the shop to be verified. The shop name analysis result may be obtained by processing the second text and the second image using the second multimodal model as described in the above embodiment.
And obtaining a shop address analysis result aiming at the shop to be verified. The shop address analysis result may be obtained by processing the third text and the third image using the third multimodal model as described in the above embodiment.
And adjusting the first credibility according to a first matching degree between the shop name analysis result and the industry analysis result and a second matching degree between the shop address analysis result and the industry analysis result to obtain an adjusted first credibility. In general, the first matching degree and the second matching degree are both proportional to the adjusted first reliability.
Correspondingly, the determining whether the first reliability is greater than a first threshold may specifically include: and judging whether the adjusted first credibility is larger than a first threshold value. If the adjusted first reliability is greater than the first threshold, a result indicating that the preset industry information in the registration information to be verified passes verification can be generated.
When the preset store name information in the registration information to be verified is verified, after determining the second credibility of the preset store name information, the method may further include:
and acquiring an industry analysis result aiming at the shop to be verified. The industry analysis result may be obtained by processing the first text and the first image using the first multi-modal model as described in the above embodiments.
And obtaining a shop address analysis result aiming at the shop to be verified. The shop address analysis result may be obtained by processing the third text and the third image using the third multimodal model as described in the above embodiment.
And adjusting the second credibility according to a third matching degree between the industry analysis result and the shop name analysis result and a fourth matching degree between the shop address analysis result and the shop name analysis result to obtain an adjusted second credibility. In general, the third matching degree and the fourth matching degree are both proportional to the adjusted second reliability.
The determining whether the second reliability is greater than a second threshold may specifically include: and judging whether the adjusted second credibility is larger than a second threshold value. If the adjusted second credibility is greater than the second threshold, a result indicating that the preset store name information in the registration information to be verified passes verification can be generated.
When the verification is performed on the preset store address information in the registration information to be verified, after determining the third credibility of the preset store address information, the method may further include:
and acquiring an industry analysis result aiming at the shop to be verified. The industry analysis result may be obtained by processing the first text and the first image using the first multi-modal model as described in the above embodiments.
And obtaining a shop name analysis result aiming at the shop to be verified. The shop name analysis result may be obtained by processing the second text and the second image using the second multimodal model as described in the above embodiment.
And adjusting the third reliability according to a fifth matching degree between the industry analysis result and the shop address analysis result and a sixth matching degree between the shop name analysis result and the shop address analysis result to obtain an adjusted third reliability. In general, the fifth matching degree and the sixth matching degree are both proportional to the adjusted third reliability.
The determining whether the third reliability is greater than a third threshold may specifically include: and judging whether the adjusted third credibility is larger than a third threshold value. If the adjusted third reliability is greater than the third threshold, a result indicating that the preset store address information in the registration information to be verified passes verification may be generated.
In the embodiment of the present specification, since the third image may further include: street view images around the shop to be verified; in practical application, street view images around the address corresponding to the preset store address information can be obtained from a map application, so that the reliability of the preset store address information can be adjusted according to the similarity between the two street view images.
Specifically, before determining whether the third reliability is greater than a third threshold, the method may further include:
and acquiring a reference street view image at the geographic position indicated by the shop address analysis result.
And adjusting the third reliability according to the similarity between the reference street view image and the street view image in the third image. In general, the similarity between the reference street view image and the street view image in the third image is proportional to the adjusted third confidence level.
Subsequently, after the adjusted third credibility is determined to be greater than the third threshold, a result indicating that the verification is passed for the preset store address information in the registration information to be verified can be generated.
Compared with the traditional scheme of verifying the store registration information by uploading a business license and a store entrance, the verification scheme of the store registration information provided by the embodiment of the specification can obtain the verification result by enabling a merchant to shoot and upload the store video data and carrying out automatic intelligent analysis and judgment on the store video data, so that the merchant has higher counterfeiting cost and low verification cost.
The video data of the shop comprise richer shop related information such as the shop interior scene, the surrounding street scene, the shop price list and the like, and the multimodal information is fused through the multimodal model, so that the obtained information amount can be improved to the greatest extent, the most accurate analysis result can be obtained, and the problem that the algorithm misjudges the black box can be solved because the shop video data can explain the basis and the reason of the analysis result.
Meanwhile, the shop registration information verification scheme provided in the embodiment of the description can also perform registration information verification for vendors who do not have sales licenses and have no clear shop headings, so as to cover more types of merchants.
Based on the same idea, the embodiment of the present specification further provides a device corresponding to the above method.
Fig. 3 is a schematic structural diagram of a store registration information verification apparatus corresponding to fig. 2 according to an embodiment of the present specification. As shown in fig. 3, the apparatus may include:
the acquisition module 302 is used for acquiring shop video data of shops to be verified; the shop to be verified is a shop with registration information to be verified.
A first extracting module 304, configured to extract text information related to the registration information to be verified from the shop video data.
A second extracting module 306, configured to extract image information related to the registration information to be verified from the shop video data.
And the analysis module 308 is configured to process the text information and the image information by using a multi-modal model to obtain a registration information analysis result for the store to be verified.
The verification module 310 is configured to perform consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result for the to-be-verified registration information.
In the embodiment of the specification, the text information may include a first text related to the business scope of the shop; the image information may include a first image related to the business range of the shop.
Correspondingly, the analysis module 308 may be specifically configured to:
and processing the first text and the first image by using a first multi-modal model to obtain an industry analysis result for the shop to be verified.
The verification module 310 may specifically include:
and the first judging unit is used for judging whether the industry analysis result and the preset industry information in the to-be-verified registration information belong to the same industry or not to obtain a first judgment result.
And the first verification result generation unit is used for generating a verification result aiming at the preset industry information according to the first judgment result.
The first determining unit may specifically include:
and the credibility determining subunit is used for determining a first credibility that the industry analysis result and the preset industry information in the to-be-verified registration information belong to the same industry according to preset industry classification information.
And the judging subunit is used for judging whether the first credibility is greater than a first threshold value.
The apparatus in fig. 3, the verification module 310, may further include:
a first acquisition unit configured to acquire a store name analysis result for the store to be authenticated.
And the second acquisition unit is used for acquiring a shop address analysis result aiming at the shop to be verified.
And the first adjusting unit is used for adjusting the first credibility according to a first matching degree between the shop name analysis result and the industry analysis result and a second matching degree between the shop address analysis result and the industry analysis result to obtain an adjusted first credibility.
The determining subunit may be specifically configured to: and judging whether the adjusted first credibility is larger than a first threshold value.
In an embodiment of the present description, the first text may include at least one of a store signboard text and a store tariff text; the first image may include at least one of an exterior view image and an interior view image of the store.
In the embodiment of the specification, the text information may further include a second text related to the shop name; the image information may further include a second image related to the shop name.
Correspondingly, the analysis module 308 may be specifically configured to:
and processing the second text and the second image by using a second multi-modal model to obtain a shop name analysis result for the shop to be verified.
The verification module 310 may specifically include:
and the second credibility determining unit is used for determining the second credibility of the preset shop name information according to the text similarity between the shop name analysis result and the preset shop name information in the to-be-verified registration information.
And the second judging unit is used for judging whether the second credibility is greater than a second threshold value or not to obtain a second judgment result.
And the second verification result generation unit is used for generating a verification result aiming at the preset shop name information according to the second judgment result.
The verification module 310 may further include:
and the third acquisition unit is used for acquiring an industry analysis result aiming at the shop to be verified.
And the fourth acquisition unit is used for acquiring a shop address analysis result aiming at the shop to be verified.
And the second adjusting unit is used for adjusting the second credibility according to a third matching degree between the industry analysis result and the shop name analysis result and a fourth matching degree between the shop address analysis result and the shop name analysis result to obtain an adjusted second credibility.
The second determining unit may be specifically configured to: and judging whether the adjusted second credibility is larger than a second threshold value.
In this specification embodiment, the second text may include store signboard text; the second image may comprise a store signboard image.
In the embodiment of the present specification, the text information may further include a third text related to the address of the store; the image information may further include a third image related to the shop address.
Correspondingly, the analysis module 308 may be specifically configured to:
and processing the third text and the third image by using a third multi-modal model to obtain a shop address analysis result for the shop to be verified.
The verification module 310 may specifically include:
and the third credibility determining unit is used for determining the third credibility of the preset store address information according to the distance value between the geographic position indicated by the store address analysis result and the geographic position indicated by the preset store address information in the to-be-verified registration information.
And the third judging unit is used for judging whether the third credibility is greater than a third threshold value or not to obtain a third judgment result.
And the third verification result generation unit is used for generating a verification result aiming at the preset shop address information according to the third judgment result.
The verification module 310 may further include:
and the fifth acquisition unit is used for acquiring the industry analysis result of the shop to be verified.
A sixth acquisition unit configured to acquire a store name analysis result for the store to be authenticated.
And the third adjusting unit is used for adjusting the third reliability according to a fifth matching degree between the industry analysis result and the shop address analysis result and a sixth matching degree between the shop name analysis result and the shop address analysis result to obtain an adjusted third reliability.
The third determining unit may be specifically configured to: and judging whether the adjusted third credibility is larger than a third threshold value.
In an embodiment of the present specification, the third image may further include: street view image.
Correspondingly, the verification module 310 may further include:
and a seventh obtaining unit, configured to obtain a reference street view image at the geographic position indicated by the store address analysis result.
And the fourth adjusting unit is used for adjusting the third credibility according to the similarity between the reference street view image and the street view image in the third image.
In an embodiment of the present specification, the third text may include: at least one of a store house sign text and a street sign text; the third image may include: at least one of a store doorplate image, a street sign image, and a street view image.
Based on the same idea, the embodiment of the present specification further provides a device corresponding to the above method.
Fig. 4 is a schematic structural diagram of a store registration information verification apparatus corresponding to fig. 2 provided in an embodiment of the present specification. As shown in fig. 4, the apparatus 400 may include:
at least one processor 410; and the number of the first and second groups,
a memory 430 communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory 430 stores instructions 420 executable by the at least one processor 410 to enable the at least one processor 410 to:
obtaining shop video data of a shop to be verified; the shop to be verified is a shop with registration information to be verified.
And extracting text information related to the to-be-verified registration information from the shop video data.
And extracting image information related to the to-be-verified registration information from the shop video data.
And processing the text information and the image information by utilizing a multi-modal model to obtain a registration information analysis result aiming at the shop to be verified.
And performing consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result aiming at the to-be-verified registration information.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus shown in fig. 4, since it is substantially similar to the method embodiment, the description is simple, and the relevant points can be referred to the partial description of the method embodiment.
In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital character system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate a dedicated integrated circuit chip. Furthermore, nowadays, instead of manually making an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Hardware Description Language), traffic, pl (core universal Programming Language), HDCal (jhdware Description Language), lang, Lola, HDL, laspam, hardward Description Language (vhr Description Language), vhal (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims (28)

1. A store registration information verification method includes:
obtaining shop video data of a shop to be verified; the shop to be verified is a shop with registration information to be verified;
extracting text information related to the to-be-verified registration information from the shop video data;
extracting image information related to the to-be-verified registration information from the shop video data;
processing the text information and the image information by using a multi-modal model to obtain a registration information analysis result for the shop to be verified;
and performing consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result aiming at the to-be-verified registration information.
2. The method of claim 1, wherein the text information comprises a first text related to a shop operating area; the image information comprises a first image related to the business range of the shop;
the processing the text information and the image information by using the multi-modal model to obtain the registration information analysis result for the shop to be verified specifically comprises:
and processing the first text and the first image by using a first multi-modal model to obtain an industry analysis result for the shop to be verified.
3. The method according to claim 2, wherein the performing consistency verification on the registration information analysis result and the registration information to be verified to obtain a verification result for the registration information to be verified specifically includes:
judging whether the industry analysis result and preset industry information in the to-be-verified registration information belong to the same industry or not to obtain a first judgment result;
and generating a verification result aiming at the preset industry information according to the first judgment result.
4. The method according to claim 3, wherein the determining whether the industry analysis result and the preset industry information in the registration information to be verified belong to the same industry specifically comprises:
determining a first credibility that the industry analysis result and preset industry information in the to-be-verified registration information belong to the same industry according to preset industry classification information;
and judging whether the first credibility is larger than a first threshold value.
5. The method of claim 4, before determining whether the first confidence level is greater than a first threshold, further comprising:
obtaining a shop name analysis result aiming at the shop to be verified;
obtaining a store address analysis result aiming at the store to be verified;
adjusting the first credibility according to a first matching degree between the shop name analysis result and the industry analysis result and a second matching degree between the shop address analysis result and the industry analysis result to obtain an adjusted first credibility;
the determining whether the first reliability is greater than a first threshold specifically includes:
and judging whether the adjusted first credibility is larger than a first threshold value.
6. The method of claim 2, the first text comprising at least one of a store sign text and a store tariff text; the first image includes at least one of an exterior view image and an interior view image of the store.
7. The method of claim 1, wherein the text message includes a second text associated with a store name; the image information comprises a second image related to the shop name;
the processing the text information and the image information by using the multi-modal model to obtain the registration information analysis result for the shop to be verified specifically comprises:
and processing the second text and the second image by using a second multi-modal model to obtain a shop name analysis result for the shop to be verified.
8. The method according to claim 7, wherein the performing consistency verification on the registration information analysis result and the registration information to be verified to obtain a verification result for the registration information to be verified specifically includes:
determining a second credibility of the preset shop name information according to the text similarity between the shop name analysis result and the preset shop name information in the to-be-verified registration information;
judging whether the second credibility is greater than a second threshold value or not to obtain a second judgment result;
and generating a verification result aiming at the preset shop name information according to the second judgment result.
9. The method of claim 8, before determining whether the second confidence level is greater than a second threshold, further comprising:
acquiring an industry analysis result for the shop to be verified;
obtaining a store address analysis result aiming at the store to be verified;
adjusting the second credibility according to a third matching degree between the industry analysis result and the shop name analysis result and a fourth matching degree between the shop address analysis result and the shop name analysis result to obtain an adjusted second credibility;
the determining whether the second reliability is greater than a second threshold specifically includes:
and judging whether the adjusted second credibility is larger than a second threshold value.
10. The method of claim 7, the second text comprising store signboard text; the second image comprises a store sign image.
11. The method of claim 1, wherein the text message comprises a third text associated with a store address; the image information comprises a third image related to a shop address;
the processing the text information and the image information by using the multi-modal model to obtain the registration information analysis result for the shop to be verified specifically comprises:
and processing the third text and the third image by using a third multi-modal model to obtain a shop address analysis result for the shop to be verified.
12. The method according to claim 11, wherein the performing consistency verification on the registration information analysis result and the registration information to be verified to obtain a verification result for the registration information to be verified specifically includes:
determining a third credibility of the preset shop address information according to a distance value between the geographical position indicated by the shop address analysis result and the geographical position indicated by the preset shop address information in the to-be-verified registration information;
judging whether the third reliability is greater than a third threshold value or not to obtain a third judgment result;
and generating a verification result aiming at the preset shop address information according to the third judgment result.
13. The method of claim 12, before determining whether the third confidence level is greater than a third threshold, further comprising:
acquiring an industry analysis result for the shop to be verified;
obtaining a shop name analysis result aiming at the shop to be verified;
adjusting the third reliability according to a fifth matching degree between the industry analysis result and the shop address analysis result and a sixth matching degree between the shop name analysis result and the shop address analysis result to obtain an adjusted third reliability;
the determining whether the third reliability is greater than a third threshold specifically includes:
and judging whether the adjusted third credibility is larger than a third threshold value.
14. The method of claim 12 or 13, the third image comprising: street view images;
before the determining whether the third reliability is greater than a third threshold, the method further includes:
acquiring a reference street view image at the geographic position indicated by the shop address analysis result;
and adjusting the third reliability according to the similarity between the reference street view image and the street view image in the third image.
15. The method of claim 11, the third text comprising: at least one of a store house sign text and a street sign text; the third image includes: at least one of a store doorplate image, a street sign image, and a street view image.
16. A store registration information verification apparatus comprising:
the system comprises an acquisition module, a verification module and a verification module, wherein the acquisition module is used for acquiring shop video data of shops to be verified; the shop to be verified is a shop with registration information to be verified;
the first extraction module is used for extracting text information related to the to-be-verified registration information from the shop video data;
the second extraction module is used for extracting image information related to the to-be-verified registration information from the shop video data;
the analysis module is used for processing the text information and the image information by utilizing a multi-modal model to obtain a registration information analysis result aiming at the shop to be verified;
and the verification module is used for performing consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result aiming at the to-be-verified registration information.
17. The apparatus of claim 16, wherein the text message comprises a first text related to a business area of a store; the image information comprises a first image related to the business range of the shop;
the analysis module is specifically configured to:
processing the first text and the first image by using a first multi-modal model to obtain an industry analysis result for the shop to be verified;
the verification module specifically comprises:
the first judging unit is used for judging whether the industry analysis result and preset industry information in the to-be-verified registration information belong to the same industry or not to obtain a first judging result;
and the first verification result generation unit is used for generating a verification result aiming at the preset industry information according to the first judgment result.
18. The apparatus according to claim 17, wherein the first determining unit specifically includes:
the credibility determining subunit is used for determining a first credibility that the industry analysis result and preset industry information in the to-be-verified registration information belong to the same industry according to preset industry classification information;
and the judging subunit is used for judging whether the first credibility is greater than a first threshold value.
19. The apparatus of claim 18, the authentication module, further comprising:
a first acquisition unit configured to acquire a store name analysis result for the store to be authenticated;
a second acquisition unit configured to acquire a store address analysis result for the store to be authenticated;
a first adjusting unit, configured to adjust the first reliability according to a first matching degree between the store name analysis result and the industry analysis result and a second matching degree between the store address analysis result and the industry analysis result, so as to obtain an adjusted first reliability;
the judging subunit is specifically configured to:
and judging whether the adjusted first credibility is larger than a first threshold value.
20. The apparatus of claim 17, the first text comprising at least one of a store sign text and a store tariff text; the first image includes at least one of an exterior view image and an interior view image of the store.
21. The apparatus of claim 16, wherein the text message includes a second text associated with a store name; the image information comprises a second image related to the shop name;
the analysis module is specifically configured to:
processing the second text and the second image by using a second multi-modal model to obtain a shop name analysis result for the shop to be verified;
the verification module specifically comprises:
a second credibility determining unit, configured to determine a second credibility of the preset store name information according to a text similarity between the store name analysis result and preset store name information in the to-be-verified registration information;
the second judging unit is used for judging whether the second credibility is greater than a second threshold value or not to obtain a second judgment result;
and the second verification result generation unit is used for generating a verification result aiming at the preset shop name information according to the second judgment result.
22. The apparatus of claim 21, the authentication module, further comprising:
a third acquisition unit, configured to acquire an industry analysis result for the store to be verified;
a fourth acquisition unit configured to acquire a store address analysis result for the store to be authenticated;
a second adjusting unit, configured to adjust the second reliability according to a third matching degree between the industry analysis result and the store name analysis result and a fourth matching degree between the store address analysis result and the store name analysis result, so as to obtain an adjusted second reliability;
the second determining unit is specifically configured to:
and judging whether the adjusted second credibility is larger than a second threshold value.
23. The apparatus of claim 21, the second text comprising store signboard text; the second image comprises a store sign image.
24. The apparatus of claim 16, wherein the text message comprises a third text associated with a store address; the image information comprises a third image related to a shop address;
the analysis module is specifically configured to:
processing the third text and the third image by using a third multi-modal model to obtain a shop address analysis result for the shop to be verified;
the verification module specifically comprises:
a third reliability determining unit, configured to determine a third reliability of the preset store address information according to a distance value between a geographic location indicated by the store address analysis result and a geographic location indicated by preset store address information in the to-be-verified registration information;
the third judging unit is used for judging whether the third credibility is greater than a third threshold value or not to obtain a third judging result;
and the third verification result generation unit is used for generating a verification result aiming at the preset shop address information according to the third judgment result.
25. The apparatus of claim 24, the authentication module, further comprising:
a fifth obtaining unit, configured to obtain an industry analysis result for the store to be verified;
a sixth acquisition unit configured to acquire a store name analysis result for the store to be authenticated;
a third adjusting unit, configured to adjust the third reliability according to a fifth matching degree between the industry analysis result and the store address analysis result and a sixth matching degree between the store name analysis result and the store address analysis result, so as to obtain an adjusted third reliability;
the third determining unit is specifically configured to:
and judging whether the adjusted third credibility is larger than a third threshold value.
26. The apparatus of claim 24 or 25, the third image comprising: street view images;
the verification module further comprises:
a seventh acquiring unit configured to acquire a reference street view image at the geographic position indicated by the shop address analysis result;
and the fourth adjusting unit is used for adjusting the third credibility according to the similarity between the reference street view image and the street view image in the third image.
27. The apparatus of claim 24, the third text comprising: at least one of a store house sign text and a street sign text; the third image includes: at least one of a store doorplate image, a street sign image, and a street view image.
28. A store registration information verification apparatus comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
obtaining shop video data of a shop to be verified; the shop to be verified is a shop with registration information to be verified;
extracting text information related to the to-be-verified registration information from the shop video data;
extracting image information related to the to-be-verified registration information from the shop video data;
processing the text information and the image information by using a multi-modal model to obtain a registration information analysis result for the shop to be verified;
and performing consistency verification on the registration information analysis result and the to-be-verified registration information to obtain a verification result aiming at the to-be-verified registration information.
CN202011572760.1A 2020-12-24 2020-12-24 Shop registration information verification method, device and equipment Active CN112613891B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011572760.1A CN112613891B (en) 2020-12-24 2020-12-24 Shop registration information verification method, device and equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011572760.1A CN112613891B (en) 2020-12-24 2020-12-24 Shop registration information verification method, device and equipment

Publications (2)

Publication Number Publication Date
CN112613891A true CN112613891A (en) 2021-04-06
CN112613891B CN112613891B (en) 2023-10-03

Family

ID=75248017

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011572760.1A Active CN112613891B (en) 2020-12-24 2020-12-24 Shop registration information verification method, device and equipment

Country Status (1)

Country Link
CN (1) CN112613891B (en)

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150347857A1 (en) * 2014-06-03 2015-12-03 Alibaba Group Holding Limited Method and system for verifying identities
US20170244908A1 (en) * 2016-02-22 2017-08-24 GenMe Inc. Video background replacement system
US20170243058A1 (en) * 2014-10-28 2017-08-24 Watrix Technology Gait recognition method based on deep learning
CN107341805A (en) * 2016-08-19 2017-11-10 北京市商汤科技开发有限公司 Background segment and network model training, image processing method and device before image
US20180122114A1 (en) * 2016-08-19 2018-05-03 Beijing Sensetime Technology Development Co., Ltd. Method and apparatus for processing video image and electronic device
CN108875676A (en) * 2018-06-28 2018-11-23 北京旷视科技有限公司 Biopsy method, apparatus and system
CN109214280A (en) * 2018-07-27 2019-01-15 北京三快在线科技有限公司 Shop recognition methods, device, electronic equipment and storage medium based on streetscape
CN109271850A (en) * 2018-08-02 2019-01-25 北京三快在线科技有限公司 A kind of Business Information method for uploading, device, electronic equipment and storage medium
CN111651674A (en) * 2020-06-03 2020-09-11 北京妙医佳健康科技集团有限公司 Bidirectional searching method and device and electronic equipment
CN111985936A (en) * 2020-08-17 2020-11-24 支付宝(杭州)信息技术有限公司 Method, device and equipment for checking merchant certificate information

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150347857A1 (en) * 2014-06-03 2015-12-03 Alibaba Group Holding Limited Method and system for verifying identities
US20170243058A1 (en) * 2014-10-28 2017-08-24 Watrix Technology Gait recognition method based on deep learning
US20170244908A1 (en) * 2016-02-22 2017-08-24 GenMe Inc. Video background replacement system
CN107341805A (en) * 2016-08-19 2017-11-10 北京市商汤科技开发有限公司 Background segment and network model training, image processing method and device before image
US20180122114A1 (en) * 2016-08-19 2018-05-03 Beijing Sensetime Technology Development Co., Ltd. Method and apparatus for processing video image and electronic device
CN108875676A (en) * 2018-06-28 2018-11-23 北京旷视科技有限公司 Biopsy method, apparatus and system
CN109214280A (en) * 2018-07-27 2019-01-15 北京三快在线科技有限公司 Shop recognition methods, device, electronic equipment and storage medium based on streetscape
CN109271850A (en) * 2018-08-02 2019-01-25 北京三快在线科技有限公司 A kind of Business Information method for uploading, device, electronic equipment and storage medium
CN111651674A (en) * 2020-06-03 2020-09-11 北京妙医佳健康科技集团有限公司 Bidirectional searching method and device and electronic equipment
CN111985936A (en) * 2020-08-17 2020-11-24 支付宝(杭州)信息技术有限公司 Method, device and equipment for checking merchant certificate information

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
邱如英, 广东经济出版社, pages: 222 *

Also Published As

Publication number Publication date
CN112613891B (en) 2023-10-03

Similar Documents

Publication Publication Date Title
US11610394B2 (en) Neural network model training method and apparatus, living body detecting method and apparatus, device and storage medium
CN107690657B (en) Trade company is found according to image
US20200364802A1 (en) Processing method, processing apparatus, user terminal and server for recognition of vehicle damage
RU2695489C1 (en) Identification of fields on an image using artificial intelligence
US11663655B2 (en) Augmenting online transaction statements using e-commerce receipts
CN109190007B (en) Data analysis method and device
CN103383732B (en) Image processing method and device
CN111311316B (en) Method and device for depicting merchant portrait, electronic equipment, verification method and system
US20190026575A1 (en) Living body detecting method and apparatus, device and storage medium
CN111783645A (en) Character recognition method and device, electronic equipment and computer readable storage medium
WO2021169529A1 (en) Method, apparatus and device for identifying risk in code image
WO2024051609A1 (en) Advertisement creative data selection method and apparatus, model training method and apparatus, and device and storage medium
CN111932363A (en) Identification and verification method, device, equipment and system for authorization book
CN113221918A (en) Target detection method, and training method and device of target detection model
CN111062770B (en) Merchant identification method, device and computer readable medium
CN110874534A (en) Data processing method and data processing device
CN109903166B (en) Data risk prediction method, device and equipment
CN110363206B (en) Clustering of data objects, data processing and data identification method
CN109087439A (en) Bill method of calibration, terminal device, storage medium and electronic equipment
CN115222464A (en) Data processing method and device, electronic equipment and storage medium
CN111985936A (en) Method, device and equipment for checking merchant certificate information
CN112613891B (en) Shop registration information verification method, device and equipment
CN111985935A (en) Method, device and equipment for checking merchant information
CN114299509A (en) Method, device, equipment and medium for acquiring information
CN113837772A (en) Method, device and equipment for auditing marketing information

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant