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

A vehicle type identification method and system based on automatic amplified samples

A vehicle recognition and sample technology, applied in the field of target recognition and vehicle recognition, can solve the problems of limited recognition rate and low recognition rate, and achieve the effect of solving difficult training.

Active Publication Date: 2022-03-29
HEFEI UNIV OF TECH
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the case of insufficient samples, the recognition rate of this system is limited to a certain extent, and the recognition rate is not high.

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
  • A vehicle type identification method and system based on automatic amplified samples
  • A vehicle type identification method and system based on automatic amplified samples
  • A vehicle type identification method and system based on automatic amplified samples

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] The present invention will be further described below in conjunction with specific embodiments.

[0040] The method described is as figure 1 As shown, the steps include:

[0041] S1: Preprocessing and labeling of vehicle images:

[0042] Control the camera device to take pictures of the vehicle in motion, manually mark and preprocess all the obtained images, mark the position and size of the bounding box where the vehicle is located, and the type of the vehicle. Normalize the size of all images to 448×448, and perform mirroring, averaging, centering, and random flipping on all images before training to increase the number of images. If there is a single-channel image, it is changed to a three-channel image.

[0043] S2: Training of the sample generation module:

[0044] Normalize the vehicle image in the bounding box in S1 to a size of 256×256, and input it into the generation confrontation network. like figure 2 As shown, the generative adversarial network consi...

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 provides a vehicle type identification method and system based on automatic amplified samples. The steps of the vehicle identification method are as follows: firstly, the vehicle image is preprocessed and marked; then the vehicle detection network is trained; at the same time, the sample generation module based on the generative confrontation network is used to generate new vehicle image data, which is the generated sample, and the generated sample is automatically Annotate, train the model recognition network; finally deploy the online vehicle detection network and model recognition network. The system includes a vehicle image preprocessing and labeling module, a vehicle detection module, a sample generation module, an automatic labeling module for generated samples and a vehicle type identification module. The car model recognition method solves the problem that the car model recognition rate is not high for vehicle images taken from various angles in the case of insufficient samples.

Description

technical field [0001] The invention relates to the technical fields of target recognition and vehicle recognition, in particular to a method and system for recognizing a vehicle model based on automatic sample amplification. Background technique [0002] In today's society, with the improvement of people's living standards and the accelerated development of urbanization, the number of vehicles per capita increases accordingly, and the resulting traffic and social problems are also becoming more and more prominent. Vehicle recognition technology is an important branch of computer vision and intelligent transportation. It has a wide range of applications in the fields of analyzing traffic flow, regulating traffic order, parking lot fee management, bayonet system, traffic accident detection and combating stolen vehicles. [0003] The application date is December 30, 2014, the publication date is July 19, 2019, and the authorized announcement number is CN105809088B for the Chin...

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 Patents(China)
IPC IPC(8): G06V20/54G06N3/04G06N3/08
CPCG06N3/08G06V20/54G06V20/52G06V2201/08G06N3/045
Inventor 余烨杨昌东路强陈维笑程茹秋
Owner HEFEI UNIV OF TECH
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