Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Vehicle identity recognition model construction method and system based on deep learning and vehicle identity recognition method and system based on deep learning

A technology of vehicle identification and deep learning, which is applied in the field of identification and vehicle identification model construction, can solve the problems of inaccurate vehicle identification and insufficient fine-grainedness, and achieve the effect of improving feature extraction ability and optimal model expression ability

Active Publication Date: 2019-10-25
XIDIAN UNIV
View PDF6 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of the lack of fine-grained vehicle feature extraction and inaccurate vehicle identification in the prior art, the present invention proposes a vehicle identification model construction and identification method and system based on deep learning, adopting the following technical solutions accomplish:

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Vehicle identity recognition model construction method and system based on deep learning and vehicle identity recognition method and system based on deep learning
  • Vehicle identity recognition model construction method and system based on deep learning and vehicle identity recognition method and system based on deep learning
  • Vehicle identity recognition model construction method and system based on deep learning and vehicle identity recognition method and system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] This embodiment discloses a method for building a vehicle identification model based on deep learning, including the following steps:

[0060] Step 1: Collect multiple scene pictures, preprocess the scene pictures, and obtain the preprocessed picture set;

[0061] Step 2: Annotate the preprocessed picture set to obtain a label set, the labels of the label set include vehicle, vehicle face, model and vehicle identity information;

[0062] Step 3: Use the preprocessed image set and label set to train the network;

[0063] The network includes a vehicle detection network, a vehicle face component extraction network, a vehicle classification network and a vehicle identification network;

[0064] The vehicle detection network is used to output vehicle target pictures;

[0065] The vehicle face component extraction network is used to output the vehicle face component picture using the vehicle target picture;

[0066] The vehicle classification network is used to output veh...

Embodiment 2

[0086] This embodiment discloses a vehicle identification method based on deep learning, comprising the following steps:

[0087] Step 1: collect vehicle identity feature vectors output from multiple scene pictures to establish a vehicle information database;

[0088] Step 2: Obtain the picture to be identified, and use the vehicle detection network in any of the deep learning-based phased vehicle identification model construction methods in Embodiment 1 to obtain the vehicle target picture of the picture to be identified;

[0089] Step 3: Through the vehicle face part extraction network in the deep learning-based phased vehicle identity recognition model construction method, the vehicle face part picture is obtained, and the vehicle face part picture of the image to be recognized is input into the deep learning-based phased vehicle identity The vehicle classification network in the recognition model construction method obtains the vehicle classification result of the picture ...

Embodiment 3

[0097] This embodiment discloses a vehicle identity recognition model construction and recognition system based on deep learning, such as figure 2 As shown, including the vehicle face detection module, vehicle classification module and vehicle identification module;

[0098] The vehicle face detection module is used to mark the vehicle with the preprocessed scene picture, and output the vehicle target picture; this module is one of the three major modules of the system, and is the basis of subsequent module functions. It mainly realizes the detection of vehicles in the input large image and the cutting and storage of vehicles. The pre-start module is responsible for the image reading and model loading, which is an essential part of each functional module.

[0099] The vehicle type classification module includes a vehicle face part extraction submodule and a vehicle type classification submodule, and the vehicle face part extraction submodule is used to mark the vehicle face wit...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the technical field of computer vision, and relates to a vehicle identity recognition model construction and recognition method and a system based on deep learning. Firstly, large-scale road monitoring pictures are used for carrying out model training of vehicle detection, and a multi-loss function staged joint training strategy is adopted for training; then, part extraction is carried out on the detected vehicle face image, and classification is carried out through a feature extraction and fusion network or a common classification network according to the vehicle facepart extraction situation; and finally, the identity feature vector of the vehicle face is extracted and filtered by using a multi-task network, and similarity measurement is performed on the featureof the image to be analyzed and the feature vector of the image in the vehicle information base to obtain a vehicle identity recognition result. According to the deep learning network framework provided by the invention, the feature extraction capability of the network model in different aspects can be improved according to requirements, so that the optimal model expression capability is realized, and the identification feature vector with significant discrimination is convenient to extract.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and more specifically, relates to a vehicle identity recognition model construction and recognition method and system based on deep learning. Background technique [0002] In recent years, the computer vision theory based on neural network and machine learning has been redefined and developed qualitatively, breaking through the limitations of manual design and the traditional image feature extraction scheme with insufficient robustness, through multi-layer convolution pooling extraction The high-dimensional features fit a large amount of training data and regress the most suitable feature scheme to solve the problem. As a result, a large number of high-efficiency, strong applicability, and easy-to-adapt application examples have emerged in the field of computer vision. License plate recognition is one of the more mature applications with high value. It is widely used in traffic checkpoint...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V20/00G06F18/241Y02T10/40
Inventor 苗启广宋建锋权义宁杨仕琴盛立杰贾广戚玉涛张亮谢琨
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com