CN109582810A - A kind of dog dog management system and its implementation based on deep learning - Google Patents
A kind of dog dog management system and its implementation based on deep learning Download PDFInfo
- Publication number
- CN109582810A CN109582810A CN201811463752.6A CN201811463752A CN109582810A CN 109582810 A CN109582810 A CN 109582810A CN 201811463752 A CN201811463752 A CN 201811463752A CN 109582810 A CN109582810 A CN 109582810A
- Authority
- CN
- China
- Prior art keywords
- dog
- information
- deep learning
- module
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The present invention discloses a kind of dog dog management system and its implementation based on deep learning, which includes terminal, server, database;The terminal includes information acquisition module, information exchange module, Operation display module;The information acquisition module positions dog dog for acquiring dog dog feature;The information exchange module is used to the acquisition data of information acquisition module being stored in the database by server, and receives the feedback information of server;The Operation display module is for showing the received feedback information of information exchange module;The server be used for by its received data matched with the storing data of database and synchronize be stored in database;The database is previously stored with the record information of dog keeper pet dog by the server.The shortcomings that the present invention overcomes traditional not end-to-end models improves model robustness;Using depth learning technology, model algorithm accuracy rate is high, meets the application of actual scene.
Description
Technical field
The present invention relates in image identification technical field, specifically a kind of dog dog management system based on deep learning and its
Implementation method.
Background technique
Data survey report display, no matter domestic or external, the most animal of pet quantity is dog, and dog and people have day
Right proximities, can not only bring happy to people, and crucial moment can also protect owner.But exist in city a large amount of unreasonable
The crowd to walk a dog considerably increases the event that pet dog is hurted sb.'s feelings, and constitutes prestige to the living in peace of citizen, traffic, health, personal safety
The side of body.
Recently as the development of deep learning technology, image recognition accuracy rate is significantly improved, such as the standard of recognition of face
True rate is up to 99.8%.Image recognition technology based on deep learning can be automatically performed feature extraction and classification task, and having makes
With it is simple, be easy to industrialize, recognition accuracy is high the advantages that.
Tional identification multitask method multistep carries out, and total algorithm robustness is low, and accuracy rate is not high, and real-time is also difficult full
Application in sufficient actual scene;It is realized for the identification of dog dog currently without the relevant technologies and algorithm.
Summary of the invention
The purpose of the present invention is to provide a kind of dog dog management system and its implementation based on deep learning, to solve
The problems mentioned above in the background art.
In order to solve the above-mentioned technical problem, the technical solution adopted by the present invention is that:
A kind of dog dog management system based on deep learning, including terminal, server, database;
The terminal includes information acquisition module, information exchange module, Operation display module;
The information acquisition module positions dog dog for acquiring dog dog feature;
The information exchange module is used to the acquisition data of information acquisition module being stored in the data by server
Library, and receive the feedback information of server;
The Operation display module is for showing the received feedback information of information exchange module;
The server be used for by its received data matched with the storing data of database and synchronize be stored in
Database;
The database is previously stored with the record information of dog keeper pet dog by the server.
Preferably, the information acquisition module acquires several features including dog dog face, posture, hair color feature.
Preferably, the implementation method of a kind of dog dog management system based on deep learning, includes the following steps:
Step 1, the image data of target dog dog is obtained using information acquisition module, and records current information acquisition module
Location information;
Step 2, using information exchange module, the image data that will acquire is sent into trained depth in server system
It practises in model, dog dog in image is identified;
Step 3, according to the recognition result of deep learning model, and dog in background data base is fed back in Operation display module
Dog information or warning information and camera position information.
Preferably, deep learning model includes following functions module in the step 2:
(1) detection module realizes detection and positioning to dog dog in piece image, that is, judges to whether there is in input picture
Dog dog;
(2) alignment module, detection dog dog face, posture key point information are simultaneously realized face alignment and posture alignment, are improved
The accuracy rate of dog dog identification;
(3) identification module extracts dog dog feature using deep learning technology, realizes accurately identifying for dog dog.
Preferably, the training method of deep learning model is as follows in the step 2:
(1) dog dog image data is arranged, data set is manually marked, and is classified as training set data, verifying collection
Data and test set data;
(2) by picture and corresponding mark image the input deep learning model in training set data and verifying collection data into
Row training, each Optimized Iterative undated parameter, until model is restrained;
(3) final performance assessment is carried out to trained deep learning model using test set data.
The beneficial effects of the present invention are:
The present invention provides a kind of dog dog management system and its implementation based on deep learning, overcomes traditional not end-to-end
The shortcomings that model, improves model robustness;Using deep learning technology, model algorithm accuracy rate is high, and real-time is good, meets practical
The application of scene;Additional equipment is not needed to provide, user can pass through APP that mobile phone-downloaded is installed or pet shop's monitoring, cell
Monitoring, public place monitoring, public security monitoring system etc. can be carried out legacy specification and be supported dog correlation with system of the invention is applied
Application.Using mobile phone shooting, collecting or pet shop's monitoring, cell monitoring, public place monitors, public security monitoring system acquires
The characteristic information and recording information bits of dog dog are set, and server is then uploaded to, and server passes through deep learning technology, server system
The deep learning model established in system matches the characteristic information of acquisition dog dog with storage information in database, if matching
Success, then explanation is the legal dog dog placed on record, and feedback user illustrates to be unregistered illegal if matching is unsuccessful
Dog dog, and feedback user and the unrestrained dog salvage service of informing in time succour dog dog.Wherein, it is main after dog dog identifies
Be applied to: display dog owner, the age of dog dog, kind, whether certificates handling, certificate date, either with or without preventive inoculation, if be large-scale
Dog, strong type dog, if be vagrant dog etc..
Detailed description of the invention
The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
Fig. 1 is present system block diagram.
Fig. 2 is implementation method flow chart of the present invention.
Specific embodiment
Technical solution of the present invention is clearly and completely described below in conjunction with attached drawing, it is clear that described implementation
Example is a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill
Personnel's every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
As shown in Figure 1, the present invention provides a kind of dog dog management system based on deep learning, including terminal, server, number
According to library;
The terminal includes information acquisition module, information exchange module, Operation display module;
The information acquisition module positions dog dog for acquiring dog dog feature;
The information exchange module is used to the acquisition data of information acquisition module being stored in the data by server
Library, and receive the feedback information of server;
The Operation display module is for showing the received feedback information of information exchange module;
The server be used for by its received data matched with the storing data of database and synchronize be stored in
Database;
The database is previously stored with the record information of dog keeper pet dog by the server.
The information acquisition module acquires several features including dog dog face, posture, hair color feature.
A kind of implementation method of dog dog management system based on deep learning, includes the following steps:
Step 1, the image data of target dog dog is obtained using information acquisition module, and records current information acquisition module
Location information;
Step 2, using information exchange module, the image data that will acquire is sent into trained depth in server system
It practises in model, dog dog in image is identified;
Step 3, according to the recognition result of deep learning model, and dog in background data base is fed back in Operation display module
Dog information or warning information and camera position information illustrate that target dog dog is the legal dog placed on record if successful match
Dog, and feedback user illustrate that target dog dog is unregistered illegal dog dog, and feedback user and informing if matching is unsuccessful
Unrestrained dog salvage service in time succours dog dog.
Deep learning model includes following functions module in the step 2:
(1) detection module realizes detection and positioning to dog dog in piece image, that is, judges to whether there is in input picture
Dog dog;
(2) alignment module, detection dog dog face, posture key point information are simultaneously realized face alignment and posture alignment, are improved
The accuracy rate of dog dog identification;
(3) identification module extracts dog dog feature using deep learning technology, realizes accurately identifying for dog dog.
The training method of deep learning model is as follows in the step 2:
(1) dog dog image data is arranged, data set is manually marked, and is classified as training set data, verifying collection
Data and test set data;
(2) by picture and corresponding mark image the input deep learning model in training set data and verifying collection data into
Row training, each Optimized Iterative undated parameter, until model is restrained;
(3) final performance assessment is carried out to trained deep learning model using test set data.
The embodiments of the present invention described above are not intended to limit the scope of the present invention, any in the present invention
Spirit and principle within made modifications, equivalent substitutions and improvements etc., should be included in claim protection model of the invention
Within enclosing.
Claims (5)
1. a kind of dog dog management system based on deep learning, which is characterized in that including terminal, server, database;
The terminal includes information acquisition module, information exchange module, Operation display module;
The information acquisition module positions dog dog for acquiring dog dog feature;
The information exchange module is used to the acquisition data of information acquisition module being stored in the database by server, and
Receive the feedback information of server;
The Operation display module is for showing the received feedback information of information exchange module;
The server be used for by its received data matched with the storing data of database and synchronize be stored in data
Library;
The database is previously stored with the record information of dog keeper pet dog by the server.
2. a kind of dog dog management system based on deep learning according to claim 1, which is characterized in that the information is adopted
Collect several features of the module acquisition including dog dog face, posture, hair color feature.
3. a kind of implementation method of dog dog management system based on deep learning according to claim 1, which is characterized in that
Include the following steps:
Step 1, the image data of target dog dog is obtained using information acquisition module, and records the position of current information acquisition module
Information;
Step 2, using information exchange module, the image data that will acquire is sent into trained deep learning mould in server system
In type, dog dog in image is identified;
Step 3, according to the recognition result of deep learning model, and dog dog letter in background data base is fed back in Operation display module
Breath or warning information and camera position information.
4. a kind of implementation method of dog dog management system based on deep learning according to claim 3, which is characterized in that
Deep learning model includes following functions module in the step 2:
(1) detection module realizes detection and positioning to dog dog in piece image, that is, judges in input picture with the presence or absence of dog
Dog;
(2) alignment module, detection dog dog face, posture key point information simultaneously realize face alignment and posture alignment, improve dog dog
The accuracy rate of identification;
(3) identification module extracts dog dog feature using deep learning technology, realizes accurately identifying for dog dog.
5. a kind of implementation method of dog dog management system based on deep learning according to claim 3, which is characterized in that
The training method of deep learning model is as follows in the step 2:
(1) dog dog image data is arranged, data set is manually marked, and is classified as training set data, verifying collection data
With test set data;
(2) picture in training set data and verifying collection data is instructed with corresponding mark image input deep learning model
Practice, each Optimized Iterative undated parameter, until model is restrained;
(3) final performance assessment is carried out to trained deep learning model using test set data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811463752.6A CN109582810A (en) | 2018-12-03 | 2018-12-03 | A kind of dog dog management system and its implementation based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811463752.6A CN109582810A (en) | 2018-12-03 | 2018-12-03 | A kind of dog dog management system and its implementation based on deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109582810A true CN109582810A (en) | 2019-04-05 |
Family
ID=65926888
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811463752.6A Pending CN109582810A (en) | 2018-12-03 | 2018-12-03 | A kind of dog dog management system and its implementation based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109582810A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110321796A (en) * | 2019-05-30 | 2019-10-11 | 北京迈格威科技有限公司 | A kind of dog only instructs and guides method, apparatus, system and storage medium |
CN110334593A (en) * | 2019-05-28 | 2019-10-15 | 浙江泽曦科技有限公司 | Pet recognition algorithms and system |
CN110414390A (en) * | 2019-07-13 | 2019-11-05 | 恒大智慧科技有限公司 | Canine recognition methods, system and readable storage medium storing program for executing in a kind of community |
CN111275014A (en) * | 2020-02-28 | 2020-06-12 | 恒大智慧科技有限公司 | Community pet management method, community server and storage medium |
CN111447410A (en) * | 2020-03-24 | 2020-07-24 | 安徽工程大学 | Dog state identification monitoring system and method |
-
2018
- 2018-12-03 CN CN201811463752.6A patent/CN109582810A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110334593A (en) * | 2019-05-28 | 2019-10-15 | 浙江泽曦科技有限公司 | Pet recognition algorithms and system |
CN110321796A (en) * | 2019-05-30 | 2019-10-11 | 北京迈格威科技有限公司 | A kind of dog only instructs and guides method, apparatus, system and storage medium |
CN110414390A (en) * | 2019-07-13 | 2019-11-05 | 恒大智慧科技有限公司 | Canine recognition methods, system and readable storage medium storing program for executing in a kind of community |
CN111275014A (en) * | 2020-02-28 | 2020-06-12 | 恒大智慧科技有限公司 | Community pet management method, community server and storage medium |
CN111447410A (en) * | 2020-03-24 | 2020-07-24 | 安徽工程大学 | Dog state identification monitoring system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109582810A (en) | A kind of dog dog management system and its implementation based on deep learning | |
CN108830226A (en) | Intelligent campus system based on Internet of Things | |
US6975346B2 (en) | Method for suspect identification using scanning of surveillance media | |
CN107507289A (en) | A kind of mobile terminal human face identification work-attendance checking method and system | |
CN107862773A (en) | Can reservation type recognition of face gate inhibition guest system and method | |
CN207690169U (en) | It can reservation type recognition of face gate inhibition's guest system | |
CN110910549A (en) | Campus personnel safety management system based on deep learning and face recognition features | |
CN106204815A (en) | A kind of gate control system based on human face detection and recognition | |
US20200125838A1 (en) | Monitoring inmate movement with facial recognition | |
CN105844246A (en) | Face recognition and second-generation ID card identification system with single camera cabinet machine | |
CN108364374A (en) | Face access control device based on deep learning and method | |
CN108921245B (en) | Law enforcement recorder and mobile terminal reverse code scanning binding ternary intelligent link uploading system | |
CN105844247A (en) | Bi-camera cabinet machine and face recognition and second-generation ID card identification system | |
CN101976356A (en) | Face identification method for surfing identification policy, and identification system thereof | |
CN109829072A (en) | Construct atlas calculation and relevant apparatus | |
CN110443312B (en) | Urban management method and system based on human body posture | |
CN110096606B (en) | Foreign roll personnel management method and device and electronic equipment | |
CN109829995A (en) | A kind of recognition of face classroom is registered system | |
CN110930577A (en) | Method for analyzing unregistered but actually living in personnel based on entrance guard data | |
CN111639509A (en) | Action track recording and inquiring method based on face recognition | |
CN109255408A (en) | A kind of Construction Safety hidden troubles removing method for positioning and taking pictures based on electronic beacon | |
CN111385530B (en) | Intelligent camera combined encryption method and system | |
CN117612107A (en) | Method for detecting abnormal personnel in place | |
CN101847274A (en) | Security inspection door used for court | |
CN208284002U (en) | A kind of intelligence community cell gate |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190405 |
|
WD01 | Invention patent application deemed withdrawn after publication |