Method and system for intelligently identifying and verifying dogs in combination with facial features

Abstract

The invention discloses an intelligent dog identification and verification method and system in combination with facial features, and relates to the technical field of pet image processing, and the key points of the technical scheme are that the method comprises the steps: collecting and obtaining a dog image through a shooting device; performing facial feature detection on the dog image by adopting a deep learning algorithm, and judging whether the dog image contains dog facial feature information or not; performing dog facial feature alignment operation on the dog image to obtain a standardized dog facial image; performing facial feature extraction on the obtained standardized dog facial image to obtain facial feature data of the standardized dog facial image; and performing matching verification on the acquired facial feature data and sample data in a database, and finally acquiring identity information of the dog. The system is helpful for quickly completing dog certificate handling, dog immune information, dog competition, exhibition, dog annual inspection, transportation quarantine, dog death and innocent treatment, and orderly management of stray / lost dogs.

Description

Technical Field

[0001] The present invention relates to the technical field of pet image processing, and more specifically, to a method and system for intelligently identifying and verifying dogs in combination with facial features. Background Art

[0003] There are clear requirements for dog rabies immunization, cross-regional dog transportation immunization quarantine management, and harmless disposal of dead dogs. Specific responsibilities and division of labor need to be implemented, and zoonotic diseases should be prevented and controlled at the source, and the use of veterinary rabies vaccines should be actively promoted. In addition, the increasing management pressure of dog deaths and harmless disposal, dog transportation quarantine, and stray / lost dog shelter will also greatly affect urban public security management and public health prevention and control management. It is urgent to establish and improve dog file information and promote the development of dog management.

[0004] The huge stock of dogs and the long-term growth trend have brought with them increasingly prominent dog management problems. Unidentified stray dogs injuring people are common, exposing that there are still many blind spots in dog management. In addition, the random discarding of dog feces and carcasses that have not been harmlessly treated causes environmental pollution and frequent outbreaks of infectious diseases. The phenomenon of dog abuse and abandonment has been repeatedly banned. These problems have greatly affected urban management and public health safety. However, the daily supervision and management of dogs, especially urban dog management, is still insufficient. The fundamental reason is that the traditional and inefficient dog individual certification and management model has not kept up with the changes of the times, and lacks effective grip and powerful tools.

[0005] Iris recognition technology is based on the iris in the eye for identity recognition. The dog's eye structure is composed of the sclera, iris, pupil lens, retina and other parts; the iris is the ring-shaped part located between the black pupil and the white sclera, which contains many interlaced spots, filaments, coronal shapes, stripes, crypts and other detailed features, and the iris will remain unchanged throughout the life course. These features determine the uniqueness of the iris features and also determine the uniqueness of identity recognition; therefore, the iris features of the eye can be used as the dog's identity recognition object.

[0006] At the same time, the nose print of a dog is called a nose print, which is also unique, just like a human fingerprint, which is unique and stable. The nose print of a dog is also formed by natural patterns and wrinkles, and is unique because of its depth and grayscale; research shows that the nose print of each dog is unique, whether it is the same species or other animals; this means that by collecting and storing the nose prints of dogs, each dog can be accurately identified and their identity information can be established.

[0007] Current methods for identifying lost pets include implanting microchips or wearing dog tags. These methods are cumbersome and inconvenient, and there are major obstacles to registering and identifying pets. Pet owners may have some concerns about the safety of accepting microchipped pets. At the same time, pet tags may be lost, and pets cannot be well identified and retrieved.

[0008] Therefore, the present invention aims to provide a method and system for intelligently identifying dogs in combination with facial features, so as to solve the above-mentioned problems. Summary of the invention

[0009] The purpose of the present invention is to provide a method and system for intelligent identification and verification of dogs in combination with facial features. The intelligent identification method can accurately match and obtain the identity information of the dog by intelligently identifying and processing the nose prints and irises of the dog, greatly accelerating the construction process of the electronic file for dog registration. This file covers the detailed immunization records of the dog and also includes the personal information of the pet owner. The intelligent identification method is not only convenient for daily management and information query, but also plays an irreplaceable role in multiple key scenarios. For example, in terms of pet loss and recovery, it can quickly lock the identity of the pet and improve the recovery rate; in the field of pet insurance, it ensures the accurate matching of insurance services and provides more protection for pet owners; when visiting the hospital, the pet doctor can immediately obtain the health record of the pet and provide a more personalized solution for diagnosis and treatment; in addition, it is also widely used in dog competition registration, exhibition qualification review and legality verification of dog transactions, effectively maintaining the order and safety of the pet market.

[0010] The above technical objectives of the present invention are achieved through the following technical solutions: a method and system for intelligent identification and verification of dogs combined with facial features, comprising the following steps:

[0011] S1. Acquire dog images through a photographing device;

[0012] S2, using a deep learning algorithm to perform facial feature detection on the dog image to determine whether the dog image contains dog facial feature information, if yes, executing step S3; if no, continuing to execute step S1;

[0013] S3, performing a dog facial feature alignment operation on the dog image to obtain a standardized dog facial image;

[0014] S4, extracting facial features from the obtained standardized dog facial image to obtain facial feature data of the standardized dog facial image;

[0015] S5. Match and verify the acquired facial feature data with the sample data in the database, and finally obtain the dog's identity information.

[0016] The present invention is further configured as follows: the dog facial feature alignment operation performed in step S3 is to align facial feature points by affine, rotation and scaling.

[0017] The present invention is further configured as follows: the facial feature extraction in step S4 is to extract the nose print and iris features of the dog to obtain facial feature data of a standardized dog facial image for subsequent dog identity recognition and verification.

[0018] The present invention is further configured as follows: the matching verification in step S5 is performed by setting a threshold, and when the similarity between the acquired facial feature data and the sample data in the database exceeds the set threshold, the matching result is output, thereby obtaining the identity information of the dog.

[0019] The present invention also provides a dog intelligent identification and verification system combining facial features: comprising an image acquisition module, an image detection module, a facial alignment module, a feature extraction module and an identity verification module;

[0020] The image acquisition module is used to collect and acquire dog images;

[0021] The image detection module calls the dog image obtained in the image acquisition module, and determines whether the dog image contains dog facial feature information by performing facial feature detection on the dog image. If so, the dog image is sent to the facial alignment module; if not, the dog image is returned to the image acquisition module to display an image format error;

[0022] The facial alignment module receives the dog image sent by the image detection module, and obtains a standardized dog facial image by performing a dog facial feature alignment operation;

[0023] The feature extraction module calls the standardized dog facial image in the facial alignment module and extracts facial feature data of the standardized dog facial image;

[0024] The identity authentication module calls the facial feature data in the feature extraction module, matches and verifies the acquired facial feature data with the sample data in the database, and finally obtains the identity information of the dog.

[0025] The present invention also provides an electronic device for intelligent dog identification combined with facial features: comprising at least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement an intelligent dog identification and verification method combined with facial features.

[0026] The present invention also provides a computer-readable storage medium: the computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by the computer to implement a dog intelligent identification and verification method combined with facial features.

[0027] The present invention also provides an information data processing terminal: the information data processing terminal is used to implement a dog intelligent identification and verification system combined with facial features.

[0028] In summary, the present invention has the following beneficial effects:

[0029] 1. This intelligent recognition method obtains dog images and extracts facial features, then matches and verifies the extracted facial feature data with the sample data in the database. It can accurately match and obtain the dog's identity information, which helps to quickly complete the establishment of electronic dog registration files including dog immunization information, pet owner information, etc., and provides convenience for multiple application scenarios, including but not limited to efficient recovery of lost pets, accurate docking of pet insurance services, instant sharing of hospital visit information, orderly management of dog competitions and exhibitions, and transparent operation of dog transactions. At the same time, the system manages and analyzes data on the entire life cycle of dogs, records and provides a number of data query services including annual dog inspections, transportation quarantine, dog deaths and harmless disposal, and shelter management of stray / lost dogs, which helps to clarify and optimize the specific division of labor and responsibilities of relevant departments in dog management, and improve management efficiency and effectiveness;

[0030] 2. This intelligent recognition method uses geometric transformations such as affine, rotation and scaling to align the facial feature points of the dog images, so that the dog's eyes, nose and other parts are moved to the same position to eliminate the influence of factors such as the dog's facial expression and head posture, and obtain standardized dog facial images to facilitate subsequent dog identification verification;

[0031] 3. The iris features of the dog's eyes and the nose print features are the dog's unique and only identity information. The intelligent recognition method and system use the iris features of the dog's eyes and the nose print features as the dog's identity information. By performing intelligent recognition processing on the dog's nose print and iris, the dog's facial feature data can be accurately obtained, ensuring the accuracy of the dog's identity recognition and verification results, and improving the accuracy of identity recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a flowchart of a dog intelligent identification and verification method combining facial features in Example 1 of the present invention;

[0033] Figure 2It is a flow chart of a method for intelligently identifying and verifying dogs in combination with facial features in Embodiment 1 of the present invention;

[0034] Figure 3 It is a module structure diagram of a dog intelligent identification and verification system combining facial features in Example 2 of the present invention. DETAILED DESCRIPTION

[0035] The following is combined with Figure 1-3 The present invention is described in further detail.

[0036] Embodiment 1: A method for intelligent dog identification and verification based on facial features, such as Figure 1 , Figure 2 As shown, the following steps are included:

[0037] S1. Acquire the dog image by using a shooting device, wherein the shooting device uses a smart shooting device such as a mobile phone or a tablet computer to shoot the dog image;

[0038] S2, using a deep learning algorithm to perform facial feature detection on the dog image to determine whether the dog image contains dog facial feature information, if yes, then executing step S3; if no, then continuing to execute step S1, wherein the purpose of facial detection is to find the position of the dog's face in the image, so as to obtain the coordinate information of the dog's face when the dog's face is found, or to cut out the dog's face, and the specific operation method is to use a directional gradient histogram to detect the dog's face position, first gray the dog image, then calculate the gradient of the pixels in the image, and then obtain the dog's face position by transforming the image;

[0039] S3, aligning the facial feature points of the dog image by affine, rotation and scaling, that is, moving the dog's eyes, nose, mouth and other parts to the same position to obtain a standardized dog facial image;

[0040] S4, extracting nose prints and iris features of the standardized dog facial image to obtain facial feature data of the standardized dog facial image, and then converting the obtained facial feature data into a compact and discriminable feature vector;

[0041] S5. Set a threshold in advance, match and verify the feature vector of the acquired facial feature data with the feature vector of the sample data in the database. When the similarity between the acquired facial feature data and the sample data in the database exceeds the set threshold, the matching result is output to obtain the dog's identity information.

[0042] Embodiment 2: A dog intelligent identification and verification system combining facial features, such as Figure 3As shown, it includes an image acquisition module, an image detection module, a face alignment module, a feature extraction module and an identity authentication module;

[0043] The image acquisition module is used to collect and acquire dog images;

[0044] The image detection module calls the dog image obtained in the image acquisition module, and determines whether the dog image contains dog facial feature information by performing facial feature detection on the dog image. If it does, the dog image is sent to the facial alignment module; if it does not, it is returned to the image acquisition module to display an image format error;

[0045] The facial alignment module receives the dog image sent by the image detection module, and obtains a standardized dog facial image by aligning the dog's facial features;

[0046] The feature extraction module includes a dog nose print extraction unit and a dog iris extraction unit. The dog nose print extraction unit can extract the dog's unique nose print data from the standardized dog facial image. The dog iris extraction unit can extract the dog's unique iris data from the standardized dog facial image. The feature extraction module extracts the facial feature data of the standardized dog facial image by calling the standardized dog facial image in the facial alignment module, and is used for the dog identity recognition and verification in the identity verification module.

[0047] The identity authentication module calls the facial feature data in the feature extraction module, matches and verifies the acquired facial feature data with the sample data in the database, and finally obtains the dog's identity information.

[0048] The dog intelligent identification system is compatible with mobile phone user terminals, WeChat applets and PC terminals, so that the dog intelligent identification system can be used across platforms and can be applied to pet owners, pet associations and pet-related management departments, while supporting multiple operating systems and devices; WeChat applets can be used directly in WeChat without the need for separate download and installation, and have the advantages of convenience and speed. The functions and performance of the software application terminal are more powerful, and complex business logic and interactive design can be realized, as well as customized functions and designs, which can meet the needs and experiences of different users. Through real-time synchronization of data between mobile phone user terminals, WeChat applets and PC terminals, the system has a wider audience, more comprehensive applications, and is more in line with the actual needs of users.

[0049] Embodiment 3: An electronic device for intelligent dog identification in combination with facial features, comprising at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the processor, the instructions being used to be executed by the processor to implement a method for intelligent dog identification and verification in combination with facial features of embodiment 1, the method comprising: acquiring a dog image by means of a photographing device; performing facial feature detection on the dog image using a deep learning algorithm to determine whether the dog image contains dog facial feature information, and if so, aligning the dog image; if not, re-acquiring the dog image; aligning the facial feature points of the dog image by affine, rotation and scaling to obtain a standardized dog facial image; extracting the dog's nose print and iris features from the obtained standardized dog facial image to obtain facial feature data of the standardized dog facial image; setting a threshold in advance, matching and verifying the acquired facial feature data with sample data in a database, and outputting the matched result when the similarity between the acquired facial feature data and the sample data in the database exceeds the set threshold, thereby obtaining the dog's identity information.

[0050] Embodiment 4: A computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by a computer to implement a method for intelligent dog identification and verification combined with facial features in Embodiment 1, the method comprising: acquiring a dog image by means of a photographing device; performing facial feature detection on the dog image using a deep learning algorithm to determine whether the dog image contains dog facial feature information, and if so, aligning the dog image; if not, re-acquiring the dog image; aligning the facial feature points of the dog image through affine, rotation and scaling to obtain a standardized dog facial image; extracting the dog's nose print and iris features from the obtained standardized dog facial image to obtain facial feature data of the standardized dog facial image; setting a threshold in advance, matching and verifying the acquired facial feature data with sample data in a database, and outputting the matched result when the similarity between the acquired facial feature data and the sample data in the database exceeds the set threshold, thereby obtaining the dog's identity information.

[0051] Embodiment 5: An information data processing terminal, the information data processing terminal is used to implement a dog intelligent identification and verification system combined with facial features in embodiment 2, the system comprising an image acquisition module, an image detection module, a facial alignment module, a feature extraction module and an identity verification module; the image acquisition module is used to collect and obtain dog images; the image detection module calls the dog image obtained in the image acquisition module, and determines whether the dog image contains dog facial feature information by performing facial feature detection on the dog image. If so, the dog image is sent to the facial alignment module; if not, the dog image is returned to the image acquisition module to display an image format error; the facial alignment module receives the dog image sent by the image detection module, and detects the dog facial feature information through the dog facial alignment module. A feature alignment operation is performed to obtain a standardized dog facial image; the feature extraction module includes a dog nose print extraction unit and a dog iris extraction unit. The dog nose print extraction unit can extract the dog's unique nose print data from the standardized dog facial image, and the dog iris extraction unit can extract the dog's unique iris data from the standardized dog facial image. The feature extraction module calls the standardized dog facial image in the facial alignment module and simultaneously extracts the facial feature data of the standardized dog facial image for dog identity recognition and verification in the identity verification module; the identity verification module calls the facial feature data in the feature extraction module, matches and verifies the acquired facial feature data with the sample data in the database, and finally obtains the dog's identity information.

[0052] This specific embodiment is merely an explanation of the present invention and is not a limitation of the present invention. After reading this specification, those skilled in the art may make non-creative modifications to the present embodiment as needed. However, as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

Claims

1. A dog intelligent identification and verification method combining facial features, characterized by: The following steps are involved: S1. Acquire dog images through a photographing device; S2, using a deep learning algorithm to perform facial feature detection on the dog image to determine whether the dog image contains dog facial feature information, if yes, executing step S3; if no, continuing to execute step S1; S3, performing a dog facial feature alignment operation on the dog image to obtain a standardized dog facial image; S4, extracting facial features from the obtained standardized dog facial image to obtain facial feature data of the standardized dog facial image; S5. Match and verify the acquired facial feature data with the sample data in the database, and finally obtain the dog's identity information.

2. The method for intelligent dog identification and verification based on facial features according to claim 1, characterized in that: The dog facial feature alignment operation in step S3 is to align the facial feature points through affine, rotation and scaling.

3. The method for intelligent dog identification and verification based on facial features according to claim 1, characterized in that: The facial feature extraction in step S4 is to extract the nose print and iris features of the dog to obtain the facial feature data of the standardized dog facial image for subsequent dog identity recognition and verification.

4. The method for intelligent dog identification and verification based on facial features according to claim 1, characterized in that: The matching verification in step S5 is performed by setting a threshold. When the similarity between the acquired facial feature data and the sample data in the database exceeds the set threshold, the matching result is output, thereby obtaining the dog's identity information.

5. A dog intelligent identification and verification system combining facial features, characterized by: It includes an image acquisition module, an image detection module, a face alignment module, a feature extraction module and an identity authentication module; The image acquisition module is used to collect and acquire dog images; The image detection module calls the dog image obtained in the image acquisition module, and determines whether the dog image contains dog facial feature information by performing facial feature detection on the dog image. If so, the dog image is sent to the facial alignment module; if not, the dog image is returned to the image acquisition module to display an image format error; The facial alignment module receives the dog image sent by the image detection module, and obtains a standardized dog facial image by performing a dog facial feature alignment operation; The feature extraction module calls the standardized dog facial image in the facial alignment module and extracts facial feature data of the standardized dog facial image; The identity authentication module calls the facial feature data in the feature extraction module, matches and verifies the acquired facial feature data with the sample data in the database, and finally obtains the identity information of the dog.

6. An electronic device for intelligent dog identification based on facial features, characterized by: The invention comprises at least one processor; and a memory connected to at least one of the processors in communication; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement a dog intelligent identification and verification method combining facial features according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by the computer to implement a dog intelligent identification and verification method combining facial features according to any one of claims 1-4.

8. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the dog intelligent identification and verification system combined with facial features as described in claim 5.

Images