CN112214748A

CN112214748A - Identity recognition system, method and device

Info

Publication number: CN112214748A
Application number: CN202011195931.3A
Authority: CN
Inventors: 朱荣华
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Ant Shengxin (Shanghai) Information Technology Co.,Ltd.
Priority date: 2020-10-30
Filing date: 2020-10-30
Publication date: 2021-01-12

Abstract

The embodiment of the specification provides an identity recognition system, an identity recognition method and an identity recognition device, wherein the identity recognition system comprises: the server is in communication connection with the at least one client; the client is configured to receive a guide instruction submitted by a user through a guide control of the identity recognition service, start an image acquisition assembly based on the guide instruction, extract a biological characteristic image of an object to be recognized, acquired by the image acquisition assembly, and send an identity recognition model issuing request to a server; the server is configured to receive the identity recognition model issuing request and issue a first identity recognition model to the client; the client is further configured to receive the first identity recognition model, perform identity feature recognition processing on the biological feature image by using the first identity recognition model, and generate an identity recognition result of the object to be recognized.

Description

Identity recognition system, method and device

Technical Field

The embodiment of the specification relates to the technical field of computers, in particular to an identity recognition system. One or more embodiments of the present specification also relate to an identification method, an identification apparatus, a computing device, and a computer-readable storage medium.

Background

With the development of information technology, the identification and management of animal identification information are increasingly emphasized by the public, and in many animal-oriented fields or services, such as pet insurance, identification and animal identification, scientific research management, rare species tracking, etc., the identification and management of animal identification information are indispensable parts of these fields.

However, the core of identity authentication and identity information management for animals is precise identity identification for animals. Only on the basis of accurately identifying the identity of an animal, the identity authentication and identity information management of the animal can be effectively enhanced, so that better and personalized services or researches are provided for the animal, but at present, chips are mostly implanted into the animal body for identifying the identity of the animal, so that the technical difficulty is high, the user experience is poor, and the cost is high, so that a more effective method is urgently needed for solving the problems.

Disclosure of Invention

In view of this, the embodiments of the present specification provide an identity recognition system. One or more embodiments of the present disclosure also relate to an identity recognition method, an identity recognition apparatus, a computing device, and a computer-readable storage medium, so as to solve the technical deficiencies in the prior art.

According to a first aspect of embodiments herein, there is provided an identification system comprising:

the system comprises a server and at least one client, wherein the server is in communication connection with the at least one client;

the client is configured to receive a guide instruction submitted by a user through a guide control of an identity recognition service, start an image acquisition assembly based on the guide instruction, extract a biological characteristic image of an object to be recognized, acquired by the image acquisition assembly, and send an identity recognition model issuing request to the server;

the server is configured to receive the identity recognition model issuing request and issue a first identity recognition model to the client;

the client is further configured to receive the first identity recognition model, perform identity feature recognition processing on the biological feature image by using the first identity recognition model, and generate an identity recognition result of the object to be recognized.

Optionally, the client is further configured to:

receiving a guiding instruction submitted by a user through a guiding control of the identity recognition service;

starting an image acquisition assembly based on the guide instruction, and extracting a reference biological characteristic image of the target object acquired by the image acquisition assembly;

inputting the reference biological characteristic image into an identity recognition model to be trained for training to generate a second identity recognition model;

and uploading the second identity recognition model to the server for storage.

Optionally, the client is further configured to:

carrying out accuracy calculation on the identity recognition result and a predetermined target identity recognition result of the object to be recognized;

and under the condition that the calculation result does not meet a preset accuracy threshold, optimizing the first identity recognition model according to the biological feature image, and uploading the optimized first identity recognition model to the server.

Optionally, the guidance instruction includes first identification information of the user;

the client further configured to:

acquiring reference biological characteristic images of a plurality of target objects in a database;

inputting the reference biological characteristic image and the biological characteristic image into the first identity recognition model for similarity calculation;

taking a reference biological characteristic image of which the similarity calculation result output by the first identity recognition model meets a preset condition as a first target image;

acquiring reference identification information associated with the first target image, and comparing the reference identification information with first identification information of the user;

and under the condition of consistent comparison, outputting the identity information associated with the first target image as an identity recognition result.

correspondingly, the client is further configured to:

querying a database for reference biometric images of a plurality of target objects associated with the first identification information;

taking a reference biological characteristic image of which the similarity calculation result output by the first identity recognition model meets a preset condition as a second target image;

and outputting the identity information associated with the second target image as an identity recognition result.

correspondingly, the client is further configured to:

inputting the biological characteristic image into the second identity recognition model to obtain output prediction identification information;

comparing the predicted identification information with first identification information of the user;

and under the condition of consistent comparison, taking the first identification information as an identity recognition result of the object to be recognized.

Optionally, the client is further configured to:

receiving a guiding instruction submitted by a user through a guiding control of an identity recognition service, wherein the guiding instruction comprises second identification information of the user;

taking the reference biological characteristic image as a sample image, taking second identification information of the user as a sample label, inputting the second identification information into an identity recognition model to be trained, and training to generate a second identity recognition model, wherein the second identity recognition model enables the second identification information to be associated with the reference biological characteristic image;

and uploading the identification model to the server.

Optionally, the client is further configured to:

and starting an image acquisition assembly based on the guide instruction, and extracting the biological characteristic image of the object to be identified, which is acquired by the image acquisition assembly according to a preset image acquisition frequency.

Optionally, the client is further configured to:

extracting key features contained in the biological feature image, inputting the key features into a detection model, and performing qualification detection on the key features to obtain a detection result aiming at the key features and output by the detection model;

if at least one key feature exists in the detection result, generating an acquisition guide instruction corresponding to the at least one key feature which is not qualified in detection under the condition that the at least one key feature is not qualified in detection;

and sending an acquisition guiding prompt for carrying out feature acquisition on the at least one key feature which is not qualified in detection to the user by operating the acquisition guiding instruction.

Optionally, the biometric image comprises an animal nose print image and the reference biometric image comprises an animal reference nose print image.

According to a second aspect of the embodiments of the present specification, there is provided an identity recognition method applied to a client, including:

starting the image acquisition assembly based on the guide instruction, and extracting a biological characteristic image of the object to be identified, which is acquired by the image acquisition assembly;

and carrying out identity characteristic identification processing on the biological characteristic image by using an identity identification model issued by a server to generate an identity identification result of the object to be identified.

correspondingly, the performing the identity recognition processing on the biological characteristic image by using the identity recognition model to generate the identity recognition result of the object to be recognized includes:

carrying out similarity calculation on the reference biological characteristic image and the identity recognition model issued by the biological characteristic image input server;

taking a reference biological characteristic image of which the similarity calculation result output by the identity recognition model meets a preset condition as a first target image;

correspondingly, the generating the identification result of the object to be identified by performing the identification process on the biological characteristic image by using the identification model issued by the server includes:

taking the reference biological characteristic image of which the similarity calculation result output by the identity recognition model meets the preset condition as a second target image;

Optionally, the identity recognition model is trained by:

and inputting the reference biological characteristic image into an identity recognition model to be trained for training to generate the identity recognition model.

Optionally, the identity recognition method further includes:

and uploading the identity recognition model and the reference biological characteristic image to the server for storage.

Optionally, after the extracting the biometric image of the object to be identified acquired by the image acquisition component, the method further includes:

and sending a reference biological characteristic image acquisition request and an identity recognition model issuing request to the server.

Optionally, the performing, by using an identity recognition model delivered by a server, identity feature recognition processing on the biometric image to generate an identity recognition result of the object to be recognized includes:

screening target reference biological characteristic images of a plurality of target objects associated with the first identification information from the reference biological characteristic images returned by the server;

receiving an identity recognition model issued by the server, and inputting the target reference biological characteristic image and the biological characteristic image into the identity recognition model;

taking the reference biological characteristic image of which the similarity calculation result output by the identity recognition model meets the preset condition as a third target image;

and outputting the identity information associated with the third target image as an identity recognition result.

inputting the biological characteristic image into the identity recognition model issued by the server to obtain output prediction identification information;

Optionally, the identity recognition model is trained by:

and taking the reference biological characteristic image as a sample image, taking second identification information of the user as a sample label, inputting the second identification information into an identity recognition model to be trained, and training to generate an identity recognition model, wherein the identity recognition model enables the second identification information to be associated with the reference biological characteristic image.

Optionally, the starting the image capturing component based on the guiding instruction and extracting the biometric image of the object to be identified captured by the image capturing component includes:

According to a third aspect of the embodiments of the present specification, there is provided an identity recognition apparatus, applied to a client, including:

the receiving module is configured to receive a guiding instruction submitted by a user through a guiding control of the identity recognition service;

the extraction module is configured to start the image acquisition component based on the guide instruction and extract a biological characteristic image of the object to be identified acquired by the image acquisition component;

and the identification module is configured to perform identity characteristic identification processing on the biological characteristic image by using an identity identification model issued by the server to generate an identity identification result of the object to be identified.

According to a fourth aspect of embodiments herein, there is provided a computing device comprising:

a memory and a processor;

the memory is configured to store computer-executable instructions and the processor is configured to execute the computer-executable instructions to implement the steps of the identification method.

According to a fifth aspect of embodiments herein, there is provided a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, perform the steps of the identification method.

In one embodiment of the description, a client receives a guide instruction submitted by a user through a guide control of an identity recognition service, starts an image acquisition assembly based on the guide instruction, extracts a biological characteristic image of an object to be recognized, acquired by the image acquisition assembly, and sends an identity recognition model issuing request to a server; the server receives the identity recognition model issuing request and issues a first identity recognition model to the client; the client receives the first identity recognition model, and performs identity feature recognition processing on the biological feature image by using the first identity recognition model to generate an identity recognition result of the object to be recognized;

by adopting the mode, the identity of the object to be recognized is recognized, the feasibility of an identity recognition scheme is improved, and the identity information is recognized in a biological characteristic image acquisition and identity characteristic recognition mode, so that the safety of the object to be recognized is guaranteed, the identity of the object to be recognized can be recognized at any time and any place, and the convenience is improved.

Drawings

FIG. 1 is a schematic diagram of an identification system provided in one embodiment of the present description;

FIG. 2 is a flow chart of a process of a method for identification according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of an image capture provided by an embodiment of the present description;

FIG. 4 is an interaction diagram of a method for identifying an identity according to an embodiment of the present disclosure;

FIG. 5 is a schematic view of an identification device provided in an embodiment of the present disclosure;

fig. 6 is a block diagram of a computing device according to an embodiment of the present disclosure.

Detailed Description

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present description. This description may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, as those skilled in the art will be able to make and use the present disclosure without departing from the spirit and scope of the present disclosure.

The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present specification refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It will be understood that, although the terms first, second, etc. may be used herein in one or more embodiments to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first can also be referred to as a second and, similarly, a second can also be referred to as a first without departing from the scope of one or more embodiments of the present description. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.

First, the noun terms to which one or more embodiments of the present specification relate are explained.

javascript: a development language for a front end.

tensoflow: an artificial intelligence framework realized by using a javascript language can be used in the fields of multiple machine learning and deep learning such as speech recognition or image recognition.

Nasal wrinkles: the unique lines on the nose are similar to human fingerprints, and the nose lines also have uniqueness and invariance, so that the nose lines can be used as a secret key for pet identity authentication.

In the present specification, an identification system is provided, and the present specification relates to an identification method, an identification apparatus, a computing device, and a computer-readable storage medium, which are described in detail in the following embodiments one by one.

As the authentication of the identity and the management of the identity information of animals are more and more regarded by the public, and in many animal-oriented fields or services, such as pet insurance, identification of animal authentication, scientific management, tracking of rare species, and the like, the authentication of the identity and the management of the identity information of animals are indispensable parts of these fields.

The core of carrying out authentication and identity information management to the animal carries out accurate identification to the animal, but at present for the realization carries out identification to the animal, implants the chip in the animal body mostly, and this kind of mode not only the technical difficulty is big, and user experience is poor, and the cost is higher moreover.

Based on this, the embodiment of the present specification provides an identity recognition system, which includes a server and at least one client, where the server is in communication connection with the at least one client; the client is used for receiving a guide instruction submitted by a user through a guide control of the identity recognition service, starting an image acquisition assembly based on the guide instruction, extracting a biological characteristic image of an object to be recognized, acquired by the image acquisition assembly, and sending an identity recognition model issuing request to the server; the server is used for receiving the identity recognition model issuing request and issuing a first identity recognition model to the client; the client is further used for receiving the first identity recognition model, performing identity feature recognition processing on the biological feature image by using the first identity recognition model, and generating an identity recognition result of the object to be recognized.

Fig. 1 is a schematic diagram illustrating an identification system provided in accordance with an embodiment of the present disclosure, including:

a server 102 and at least one client 104, wherein the server 102 is connected with the at least one client 104 in a communication way;

the client 104 is configured to receive a guidance instruction submitted by a user through a guidance control of an identity recognition service, start an image acquisition component based on the guidance instruction, extract a biological feature image of an object to be recognized acquired by the image acquisition component, and send an identity recognition model issuing request to the server 102;

the server 102 is configured to receive the identification model issuing request and issue a first identification model to the client 104;

the client 104 is further configured to receive the first identity recognition model, perform identity feature recognition processing on the biometric image by using the first identity recognition model, and generate an identity recognition result of the object to be recognized.

Optionally, the client 104 is further configured to:

and uploading the second identification model to the server 102 for storage.

Optionally, the client 104 is further configured to:

and under the condition that the calculation result does not meet a preset accuracy threshold, optimizing the first identity recognition model according to the biological feature image, and uploading the optimized first identity recognition model to the server 102.

the client 104, further configured to:

accordingly, the client 104 is further configured to:

Optionally, the client 104 is further configured to:

uploading the identification model to the server 102.

Optionally, the client 104 is further configured to:

The embodiment of the specification identifies the identity of the object to be identified by the above method, so that the feasibility of an identity identification scheme is improved, and the identity information is identified by the biological characteristic image acquisition and identity characteristic identification method, so that the safety of the object to be identified is guaranteed, the identity of the object to be identified can be identified at any time and any place, and the convenience is improved.

Fig. 2 shows a process flow diagram of an identity recognition method provided in accordance with an embodiment of the present specification, applied to a client, including steps 202 to 206.

Step 202, receiving a guiding instruction submitted by a user through a guiding control of an identification service.

In particular, the identification methods provided in the examples herein can be used to identify animals, including but not limited to wild animals or livestock. Feeding, domesticating or breeding; livestock includes animals fed or raised by users, including animal pets (pet dogs, pet cats, pet pigs, etc.), tea pet pets (bombesi, fabusia, etc.), and other pets (woodchuck, rabbits, hamsters, hedgehog, bats, etc.), etc., and in addition to the above-mentioned pets, livestock also includes poultry animals raised in livestock industry, such as chickens, ducks, etc., or animals raised in animal husbandry, such as cows, sheep, horses, etc.

The identification service may be an application program for identification of the client, or may be a sub-application installed in an application program and capable of providing the identification service, and may be determined specifically according to actual needs, which is not limited herein.

In the identity recognition method provided in the embodiment of the present specification, in a case that archive information of the object to be recognized is stored in a database in advance, a user may upload a biometric image of the object to be recognized through the identity recognition service to implement identity recognition on the object to be recognized, for example, in a claim learning scene, if the user makes a claim settlement request for an animal, a claim settlement service provider may perform identity recognition on the animal by acquiring a biometric image of the animal to determine whether the animal participates in a claim or meets a claim settlement condition, and the like; in the scene of finding the lost pet, the service provider can identify the pet by collecting the biometric image of the pet, so as to judge whether the pet is lost by the lost pet.

Specifically, a user clicks a guide control of the animal identification service displayed by the trigger client to submit a guide instruction, the client comprises a mobile phone, a tablet or a computer and the like, the guide instruction is used for providing a guide service for the animal identification process, and specifically can be used for triggering the client to start the image acquisition assembly to acquire a biological characteristic image, and can provide an image acquisition strategy of the biological characteristic image for the user in the process of starting the image acquisition assembly to acquire the biological characteristic image at the client.

And 204, starting the image acquisition assembly based on the guide instruction, and extracting the biological characteristic image of the object to be identified acquired by the image acquisition assembly.

Specifically, the object to be identified may include an animal whose identity needs to be identified; the image acquisition component is an image acquirer configured at a client, the image acquirer with high use frequency in daily life comprises video input equipment, namely a camera, and the image acquisition component of the client in the embodiment of the specification is a camera configured at a mobile phone, a tablet or a computer of a user; and after receiving the guide instruction, the client starts the camera to acquire images and extracts the biological characteristic image of the object to be identified acquired by the camera.

In addition, in the embodiment of the specification, the object to be identified is an feeding pet, and when the feeding pet service mechanism receives a service request sent by a user for the feeding pet, specifically during the processes of feeding, compensating, or medical diagnosis for the feeding pet, if the pre-established archive information of the feeding pet is stored in the database, the feeding pet service mechanism needs to identify the feeding pet, so as to provide corresponding service according to the identification result, for example, after the user sends the compensation request for the feeding pet, the compensating pet service mechanism needs to identify the feeding pet, so as to determine whether the pet is already fed back.

The nose print can uniquely identify the identity information of the animal, so that in order to ensure the accuracy of the identity identification result and improve the identity identification efficiency, the biological characteristic image described in the embodiment of the specification comprises an animal nose print image, correspondingly, the object to be identified described in the embodiment of the specification is an animal with the nose print, such as a pet cat, a pet dog and the like, for other types of animals, such as orangutan, panda, pig, cow, sheep, chicken, duck and the like, a facial image or an iris image can be adopted for processing, the specific processing mode is similar to the processing mode of the animal nose print image, and details are not repeated here.

In a specific implementation, after the image capturing component is started, the image capturing component may capture a biometric image according to a preset image capturing frequency, so that the image capturing component is started based on the guiding instruction, and a biometric image of the object to be recognized captured by the image capturing component is extracted, that is, the image capturing component is started based on the guiding instruction, and a biometric image of the object to be recognized captured, which is captured by the image capturing component according to the preset image capturing frequency, is extracted, or a video including a biometric feature of the object is recorded, and a biometric image meeting a condition is extracted from the video.

In practical applications, the preset image capturing frequency may be 300 ms/time, but the specific image capturing frequency may be set according to practical requirements, and is not limited herein.

In addition, the image acquisition component is started based on the guiding instruction, and the biological characteristic image of the object to be identified acquired by the image acquisition component is extracted, which can be specifically realized by the following modes:

Specifically, the eligibility detection is performed on the key features included in the biometric image, that is, whether the key features meet the identification condition is detected, for example, in the case that the biometric image is an animal rhinoprint image, if the animal needs to be identified accurately, the acquired animal rhinoprint image needs to be a complete rhinoprint image, and therefore, in the process of performing the eligibility detection on the key features of the rhinoprint image, whether the key features are complete needs to be detected, and if the key features are missing, the detection result is not qualified.

In practical application, after the biological characteristic image of the object to be identified, which is acquired by the image acquisition assembly, is extracted, the qualification detection of the key characteristics contained in the biological characteristic image can be carried out through the detection model, so as to determine whether the acquisition guide prompt needs to be provided for a user according to the qualification detection result output by the detection model.

In the embodiment of the present description, after the key features in the biometric image are extracted, the detection model is used to obtain the qualification detection result of the key features, and if at least one key feature is unqualified in detection, an acquisition guidance instruction may be generated, where the acquisition guidance instruction may be generated for the unqualified key features in detection or generated for the entire biometric image in an acquisition manner.

After the acquisition guide instruction is generated, an acquisition guide prompt is sent to a user by operating the acquisition guide instruction, the image acquisition schematic diagram provided in the embodiment of the present description is shown in fig. 3, and the acquisition guide prompt shown in fig. 3 is "please shoot the left nose and incline to the left face by about 15 degrees". If the image acquisition component acquires the biological characteristic image according to the preset acquisition frequency, the acquired biological characteristic image can be subjected to real-time qualification detection, and the acquisition is stopped when the biological characteristic image is qualified; or, the image acquisition component may acquire the biometric images one by one, that is, perform the qualification detection on the acquired biometric images, continue to acquire if the detection is not qualified, and stop acquiring if the detection is qualified.

Further, if the acquisition guide instruction is generated for detecting unqualified key features, the user performs image acquisition again for the unqualified key features according to the acquisition guide prompt, and can integrate all the qualified key features to generate a biological feature image for identity recognition under the condition that the acquired image contains the qualified key features; if the acquisition guide instruction is generated by aiming at the acquisition mode of the whole biological characteristic image, the biological characteristic image can be used for identity recognition under the condition that the user acquires the image again according to the acquisition guide prompt and obtains the qualified biological characteristic image.

Under the condition that key features in the biological feature images are unqualified, image acquisition guiding prompts can be provided for a user, so that the user can acquire the qualified biological feature images according to a correct acquisition mode, the acquisition efficiency of the biological feature images is improved, and the identity recognition efficiency of an object to be recognized is improved.

And step 206, performing identity characteristic identification processing on the biological characteristic image by using an identity identification model issued by the server to generate an identity identification result of the object to be identified.

Specifically, the identity recognition model is uploaded to a server for storage after the client training is completed, the client can send an identity recognition model issuing request to the server under the condition that the client needs to use the identity recognition model to perform identity recognition on an object to be recognized, the server issues the identity recognition model to the client after receiving the model issuing request, and the client can input the extracted biological characteristic image into the identity recognition model to perform identity characteristic recognition processing, so that an identity recognition result of the object to be recognized is generated.

Further, after the identity recognition model is generated through training, the identity recognition model can be uploaded to the server to be stored, in addition, the reference biological characteristic image and the first identification information of the user can also be uploaded to the server to be stored, and the reference biological characteristic image, the first identification information and the identity recognition model can be used in the identity recognition process of the object to be recognized.

In practical application, in the process of identity recognition, after a biological characteristic image of an object to be recognized, which is acquired by the image acquisition assembly, is extracted, a reference biological characteristic image acquisition request and an identity recognition model issuing request are sent to the server, and the identity recognition model and the reference biological characteristic image are issued to a client by the server; or, the client may send a first identification information acquisition request to the server under the condition that the identity of the object to be recognized needs to be recognized by means of the first identification information of the user stored in the server, and the server issues the first identification information to the client.

In addition, after the client obtains the identity recognition result, the client can compare the identity recognition result with a real result (target identity recognition result) to determine the accuracy of the identity recognition result, and can optimize the identity recognition model by using the biological characteristic image and upload the optimized identity recognition model to the server under the condition that the accuracy of the identity recognition result is lower than a preset accuracy threshold.

In specific implementation, the guiding instruction comprises first identification information of the user;

correspondingly, the identity recognition model is used for carrying out identity feature recognition processing on the biological feature image to generate an identity recognition result of the object to be recognized, and the method can be specifically realized in the following way:

Specifically, the reference biometric image is a feature image of a target object acquired when a target object archive is established, and the target object archive can be established in the processes of insurance application, medical treatment and the like for the target object; in the case where the biometric image is an animal nose print image, the reference biometric image may be an animal reference nose print image.

After extracting the biological characteristic image, sending an identity recognition model issuing request and a reference biological characteristic image acquisition request to a server, to obtain a reference biological characteristic image sent by a server, and to input the reference biological characteristic image and the extracted biological characteristic image into the identity recognition model for similarity calculation, and the reference biometric image with the similarity greater than the preset similarity threshold in the calculation result is taken as a target image (first target image), and comparing the reference identification information associated with the target image with the first identification information of the user, if the comparison is consistent, the reference biological characteristic image matched with the biological characteristic image can be found in the database, and can further indicate that the user has established a pet profile for the object to be identified, and, therefore, the identity information associated with the target image can be output as an identity recognition result of the object to be recognized.

In addition, when the reference biometric image matched with the biometric image is not found in the database, it indicates that the user does not create the file of the object to be identified, and at this time, a prompting message may be sent to the user to prompt the user to create the file for the object to be identified.

The method comprises the steps of collecting a biological characteristic image through an image collection assembly of a client, calculating the similarity between the biological characteristic image and a reference biological characteristic image through an identity recognition model, using the reference biological characteristic image with the similarity calculation result meeting a preset similarity threshold as a target image, and determining an identity recognition result according to the target image.

In addition, when the guiding instruction includes the first identification information of the user, the identification model issued by the server is used to perform identification processing on the biological characteristic image to generate an identification result of the object to be identified, and the following method can be further implemented:

Specifically, after the biometric image is extracted, an identity recognition model issuing request and a reference biometric image acquisition request are sent to the server to acquire a reference biometric image associated with the first identification information issued by the server, the reference biometric image and the extracted biometric image are input to the identity recognition model for similarity calculation, then the reference biometric image with the similarity greater than a preset similarity threshold in the calculation result is used as a target image (a second target image), and the identity information associated with the target image is output as an identity recognition result.

By adopting the mode to identify the identity of the object to be identified, the feasibility of an identity identification scheme is improved, the safety of the object to be identified can be ensured, in addition, the identification process is not limited by time and place, and the convenience of the identity identification process of the object to be identified is improved.

Further, the identity recognition model is trained by the following method:

Specifically, the training process of the identity recognition model is unsupervised training, that is, only a sample image and no sample label are provided. And the process of model training is completed at the client, and under the condition that the user establishes the archive for the object to be recognized for the first time, the biological characteristic image acquired by the image acquisition assembly can be used as a reference biological characteristic image, and the reference biological characteristic image is input into the identity recognition model to be trained for training so as to generate the identity recognition model.

The embodiment of the present specification may perform model training through tensorflow, and the identity recognition model to be trained may be a neural network model, and specifically may be a fully-connected neural network.

After the model training is finished, the client side can upload the identity recognition model to a server for storage; and under the condition that other clients have the requirement of carrying out identity recognition on the object to be recognized, sending a model issuing request to the server so that the server issues the identity recognition model to the clients, after the clients recognize the biological characteristic image through the identity recognition model to obtain a recognition result, if the accuracy of the obtained recognition result is determined not to meet the condition, optimizing the identity recognition model by using the biological characteristic image, and uploading the optimized identity recognition model to the server.

The identity recognition model is optimized in the mode, and the accuracy of the output result of the model is improved.

correspondingly, the identity recognition processing is performed on the biological characteristic image by using the identity recognition model issued by the server to generate the identity recognition result of the object to be recognized, and the method can also be realized by the following steps:

Further, the identity recognition model is trained by the following method:

Specifically, the training process of the identity recognition model is supervised training, and the training process is also completed at the client, under the condition that the user establishes a file for the object to be identified for the first time, the biological characteristic image acquired by the image acquisition component can be used as a reference biological characteristic image, taking the reference biological characteristic image as a sample image, taking second identification information of the user carried in a guide instruction as a sample label, inputting the sample image and the sample label into an identity recognition model to be trained for training, to generate the identification model, the identification model associating the second identification information with the reference biometric image, namely, a biological characteristic image is input into the identity recognition model, and the user identification information related to the biological characteristic image output by the model can be obtained.

Therefore, in the identification process, after receiving a guiding instruction submitted by a user, the client can start the image acquisition assembly to acquire the biological characteristic image, and simultaneously can send an identification model issuing request to the server so as to input the acquired biological characteristic image into the identification model issued by the server for processing, thereby acquiring user identification information (predicted identification information) which is output by the model and is associated with the biological characteristic image; because the guiding instruction submitted by the user comprises the second identification information of the user, after the predicted identification information output by the model is obtained, the predicted identification information can be compared with the second identification information, and under the condition of consistent comparison, the second identification information is used as the identity recognition result of the object to be recognized.

The identity recognition method provided by the embodiment of the specification is applied to a client, and comprises the steps of receiving a guide instruction submitted by a user through a guide control of an identity recognition service, starting an image acquisition assembly based on the guide instruction, extracting a biological characteristic image of an object to be recognized, acquired by the image acquisition assembly, and performing identity recognition processing on the biological characteristic image by using an identity recognition model issued by a server to generate an identity recognition result of the object to be recognized;

by adopting the mode, the identity of the object to be recognized is recognized, the feasibility of an identity recognition scheme is improved, and the identity information of the object to be recognized is recognized by adopting the biological characteristic image acquisition and identity characteristic recognition modes, so that the safety of the object to be recognized is guaranteed, the identity of the object to be recognized can be recognized at any time and any place, and the convenience is improved.

The above is a schematic scheme of an identity recognition method of this embodiment. It should be noted that the technical solution of the identity recognition system and the technical solution of the identity recognition method belong to the same concept, and details that are not described in detail in the technical solution of the identity recognition system can be referred to the description of the technical solution of the identity recognition method.

The following description will further explain the identity recognition method by taking the application of the identity recognition method provided in this specification in a pet identity recognition scenario as an example with reference to fig. 4. Fig. 4 shows an interaction diagram of an identity recognition method provided in an embodiment of the present specification, and specific steps include steps 402 to 422.

Step 402, the client receives a guiding instruction submitted by a user through a guiding control of the pet keeping identification service.

In step 404, the client starts an image capture component based on the guiding instruction.

And 406, extracting a reference biological characteristic image of the pet, which is collected by the image collection component, by the client.

And step 408, inputting the reference biological characteristic image into the to-be-trained pet identity recognition model for training by the client, and generating the pet identity recognition model.

And step 410, uploading the feeding pet identification model to a server for storage.

Steps 402 to 410 are the training process for the pet feeding identification model.

Steps 412 to 422 are the application process of the pet feeding identification model.

In step 412, the client receives a guidance instruction submitted by the user through a guidance control of the pet keeping identification service.

And 414, starting an image acquisition component based on the guiding instruction, and extracting a biological characteristic image of the pet to be identified, which is acquired by the image acquisition component.

And step 416, the client sends an stocking pet identity recognition model issuing request to the server.

And 418, the server issues an feeding pet identity recognition model to the client.

And step 420, the client side utilizes the pet feeding identity recognition model to perform identity feature recognition processing on the biological feature image, and generates a pet feeding identity recognition result of the pet to be fed.

And step 422, the client displays the identification result of the housed pet to the user.

The embodiment of the specification identifies the identities of the feeding pets by the aid of the above manners, improves feasibility of an identity identification scheme of the feeding pets, identifies identity information of the feeding pets by means of biological characteristic image acquisition and identity characteristic identification, guarantees safety of the feeding pets, can identify the identities of the feeding pets at any time and any place, and improves convenience.

In addition, the identity recognition method can be used for tracking the action track of the wild protection animal, namely, a user can submit a guide instruction through a guide control of the wild protection animal identity recognition service, the client starts an image acquisition assembly to acquire a reference nose print characteristic image of the wild protection animal based on the guide instruction, the reference nose print characteristic image is input into a wild protection animal identity recognition model to be trained, the wild protection animal identity recognition model is generated, and the wild protection animal identity recognition model obtained through training is uploaded to a server to be stored.

Further, the client receives a guide instruction submitted by a user through a guide control of a wild protection animal identification service, starts an image acquisition assembly based on the guide instruction, extracts a nose print characteristic image of a wild protection animal to be identified, which is acquired by the image acquisition assembly, sends a request issued by a wild protection animal identification model to a server, performs identification characteristic recognition processing on the nose print characteristic image by using the wild protection animal identification model after receiving the wild protection animal identification model issued by the server, generates and displays an identification result of the wild protection animal to be identified, performs identification on the wild protection animal to be identified by the above method, improves feasibility of an identification scheme, and identifies identity information of the wild protection animal by means of the collection of the nose print characteristic image and the identification characteristic recognition, therefore, the action track of the wild protection animal can be tracked, the safety of the wild protection animal is guaranteed, the identified wild protection animal can be identified at any time and any place, and the convenience is improved.

Corresponding to the above method embodiment, the present specification further provides an identity recognition apparatus embodiment, and fig. 5 shows a schematic diagram of an identity recognition apparatus provided in an embodiment of the present specification. As shown in fig. 5, the apparatus includes:

a receiving module 502 configured to receive a guidance instruction submitted by a user through a guidance control of an identification service;

an extraction module 504 configured to start the image acquisition component based on the guiding instruction and extract a biometric image of the object to be identified acquired by the image acquisition component;

the identification module 506 is configured to perform identity feature identification processing on the biometric image by using an identity identification model issued by the server, so as to generate an identity identification result of the object to be identified.

accordingly, the identification module 506 includes:

an acquisition sub-module configured to acquire reference biometric images of a plurality of target objects in a database;

the first calculation sub-module is configured to perform similarity calculation on the reference biological characteristic image and the identity recognition model issued by the biological characteristic image input server;

a first target image determination submodule configured to take a reference biological feature image, of which a similarity calculation result output by the identity recognition model meets a preset condition, as a first target image;

a first comparison sub-module configured to acquire reference identification information associated with the first target image and compare the reference identification information with first identification information of the user;

and the first output sub-module is configured to output the identity information associated with the first target image as an identity recognition result under the condition of consistency in comparison.

accordingly, the identifying module 502 includes:

a query sub-module configured to query a database for reference biometric images of a plurality of target objects associated with the first identification information;

the second calculation sub-module is configured to perform similarity calculation on the reference biological characteristic image and the identity recognition model issued by the biological characteristic image input server;

a second target image determination submodule configured to take a reference biological feature image, of which a similarity calculation result output by the identity recognition model meets a preset condition, as a second target image;

and the second output sub-module is configured to output the identity information associated with the second target image as an identity recognition result.

Optionally, the identity recognition model is trained by:

Optionally, the identity recognition apparatus further includes:

a first upload module configured to upload the identification model and the reference biometric image to the server for storage.

Optionally, the identity recognition apparatus further includes:

the request sending module is configured to send a reference biological characteristic image obtaining request and an identification model issuing request to the server.

Optionally, the identifying module 506 includes:

a filtering sub-module configured to filter target reference biometric images of a plurality of target objects associated with the first identification information from the reference biometric images returned by the server;

the input sub-module is configured to receive an identity recognition model issued by the server and input the target reference biological characteristic image and the biological characteristic image into the identity recognition model;

a third target image determination submodule configured to take a reference biological feature image, of which a similarity calculation result output by the identity recognition model meets a preset condition, as a third target image;

a third output sub-module configured to output the identity information associated with the third target image as an identity recognition result.

accordingly, the identification module 506 includes:

the predicted identification information output sub-module is configured to input the biological characteristic image into the identity recognition model issued by the server and acquire output predicted identification information;

a second comparison sub-module configured to compare the predicted identification information with the first identification information of the user;

and the identity recognition result determining submodule is configured to take the first identification information as the identity recognition result of the object to be recognized under the condition of consistent comparison.

Optionally, the identity recognition model is trained by:

Optionally, the extracting module 504 includes:

the first extraction submodule is configured to start an image acquisition assembly based on the guiding instruction, and extract a biological characteristic image of the object to be identified, which is acquired by the image acquisition assembly according to a preset image acquisition frequency.

Optionally, the extracting module 504 includes:

the second extraction sub-module is configured to extract key features contained in the biological feature image, input the key features into a detection model, perform qualification detection on the key features, and obtain a detection result output by the detection model and aiming at the key features;

the acquisition guide instruction generation sub-module is configured to generate an acquisition guide instruction corresponding to at least one key feature which is unqualified in detection if at least one key feature exists in the detection result and is unqualified in detection;

and the acquisition guide prompt sending submodule is configured to send an acquisition guide prompt for carrying out feature acquisition on the at least one key feature which is not qualified in detection to the user by operating the acquisition guide instruction.

The above is a schematic scheme of an identification apparatus of this embodiment. It should be noted that the technical solution of the identity recognition apparatus and the technical solution of the identity recognition method belong to the same concept, and details that are not described in detail in the technical solution of the identity recognition apparatus can be referred to the description of the technical solution of the identity recognition method.

FIG. 6 illustrates a block diagram of a computing device 600 provided in accordance with one embodiment of the present description. The components of the computing device 600 include, but are not limited to, a memory 610 and a processor 620. The processor 620 is coupled to the memory 610 via a bus 630 and a database 650 is used to store data.

Computing device 600 also includes access device 640, access device 640 enabling computing device 600 to communicate via one or more networks 660. Examples of such networks include the Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the internet. Access device 640 may include one or more of any type of network interface (e.g., a Network Interface Card (NIC)) whether wired or wireless, such as an IEEE802.11 Wireless Local Area Network (WLAN) wireless interface, a worldwide interoperability for microwave access (Wi-MAX) interface, an ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a bluetooth interface, a Near Field Communication (NFC) interface, and so forth.

In one embodiment of the present description, the above-described components of computing device 600, as well as other components not shown in FIG. 6, may also be connected to each other, such as by a bus. It should be understood that the block diagram of the computing device architecture shown in FIG. 6 is for purposes of example only and is not limiting as to the scope of the present description. Those skilled in the art may add or replace other components as desired.

Computing device 600 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., tablet, personal digital assistant, laptop, notebook, netbook, etc.), mobile phone (e.g., smartphone), wearable computing device (e.g., smartwatch, smartglasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or PC. Computing device 600 may also be a mobile or stationary server.

The memory 610 is used for storing computer executable instructions, and the processor 620 is used for executing the following computer executable instructions to implement the steps of the identification method.

The above is an illustrative scheme of a computing device of the present embodiment. It should be noted that the technical solution of the computing device and the technical solution of the above-mentioned identity recognition method belong to the same concept, and details that are not described in detail in the technical solution of the computing device can be referred to the description of the technical solution of the above-mentioned identity recognition method.

An embodiment of the present specification also provides a computer readable storage medium storing computer instructions, which when executed by a processor, are used for implementing the steps of the identity recognition method.

The above is an illustrative scheme of a computer-readable storage medium of the present embodiment. It should be noted that the technical solution of the storage medium belongs to the same concept as the technical solution of the above-mentioned identification method, and details that are not described in detail in the technical solution of the storage medium can be referred to the description of the technical solution of the above-mentioned identification method.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The computer instructions comprise computer program code which may be in the form of source code, object code, an executable file or some intermediate form, or the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.

It should be noted that, for the sake of simplicity, the foregoing method embodiments are described as a series of acts, but those skilled in the art should understand that the present embodiment is not limited by the described acts, because some steps may be performed in other sequences or simultaneously according to the present embodiment. Further, those skilled in the art should also appreciate that the embodiments described in this specification are preferred embodiments and that acts and modules referred to are not necessarily required for an embodiment of the specification.

In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

The preferred embodiments of the present specification disclosed above are intended only to aid in the description of the specification. Alternative embodiments are not exhaustive and do not limit the invention to the precise embodiments described. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the embodiments and the practical application, to thereby enable others skilled in the art to best understand and utilize the embodiments. The specification is limited only by the claims and their full scope and equivalents.

Claims

1. An identification system comprising:

2. The identification system of claim 1, the client further configured to:

and uploading the second identity recognition model to the server for storage.

3. The identification system of claim 1, the client further configured to:

4. The identification system of claim 1, wherein the guiding instruction comprises first identification information of the user;

the client further configured to:

5. The identification system of claim 1, wherein the guiding instruction comprises first identification information of the user;

correspondingly, the client is further configured to:

6. The identification system of claim 1, wherein the guiding instruction comprises first identification information of the user;

correspondingly, the client is further configured to:

7. The identification system of claim 6, the client further configured to:

and uploading the identification model to the server.

8. The identification system of claim 1, the client further configured to:

9. The identification system of claim 1, the client further configured to:

10. The identification system of claim 1, the biometric image comprising an animal nose print image, the baseline biometric image comprising an animal baseline nose print image.

11. An identity recognition method is applied to a client and comprises the following steps:

12. The identification method according to claim 11, wherein the guiding instruction comprises first identification information of the user;

13. The identification method according to claim 11, wherein the guiding instruction comprises first identification information of the user;

14. The identity recognition method of claim 12 or 13, the identity recognition model being trained by:

15. The identification method of claim 14, further comprising:

16. The identity recognition method of claim 15, after the extracting the biometric image of the object to be recognized, which is acquired by the image acquisition component, further comprising:

17. The identification method according to claim 11, wherein the guiding instruction comprises first identification information of the user;

18. The identity recognition method of claim 17, the identity recognition model being trained by:

19. The identity recognition method of claim 11, wherein the starting of the image acquisition component based on the guiding instruction and the extraction of the biometric image of the object to be recognized acquired by the image acquisition component comprise:

20. The identity recognition method of claim 11, wherein the starting of the image acquisition component based on the guiding instruction and the extraction of the biometric image of the object to be recognized acquired by the image acquisition component comprise:

21. The identification method of claim 11, the biometric image comprising an animal nose print image, the baseline biometric image comprising an animal baseline nose print image.

22. An identity recognition device applied to a client comprises:

23. A computing device, comprising:

a memory and a processor;

the memory is configured to store computer-executable instructions, and the processor is configured to execute the computer-executable instructions to implement the steps of the identification method of any one of claims 11 to 21.

24. A computer readable storage medium storing computer instructions which, when executed by a processor, carry out the steps of the identification method of any one of claims 11 to 21.