A bayonet vehicle color recognition method

A color recognition and vehicle technology, applied in the field of bayonet vehicle color recognition, can solve the problems of insufficient classification and recognition accuracy, and achieve the effects of reducing color confusion, reducing the missed detection rate, and shortening the research and development cycle.

Inactive Publication Date: 2019-07-19
ZHEJIANG ICARE VISION TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Most of the above recognitions are based on experience to extract features of different color spaces and use classifiers for classification. The disadvantage is that empirical features have great limitations, resulting in insufficient classification and recognition accuracy.

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 bayonet vehicle color recognition method
  • A bayonet vehicle color recognition method
  • A bayonet vehicle color recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0017] Below in conjunction with embodiment and accompanying drawing, the present invention will be further described:

[0018] The concrete steps of this embodiment are:

[0019] 1. Using the difference between any two pixels as a feature, a vehicle detector composed of a cascaded AdaBoost classifier and a false detection classifier is used to obtain the specific position of the target vehicle. Combining very simple differences between two pixels as features through statistical learning can effectively improve the efficiency of the detector while ensuring performance.

[0020] 2. Obtain a set of clear and stable feature point positions on the front or rear of the vehicle based on target positioning, and then obtain the aligned vehicle area image through affine transformation. There are many differences in the structural information of the front and rear of the car, and the accuracy of color recognition can be effectively improved by dividing and conquering.

[0021] 3. Conv...

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 relates to a bayonet vehicle color recognition method. In the present invention, through the detection and positioning of the front and rear of the vehicle, the entire front face or the external rectangular area of ​​the rear of the vehicle is used as the region of interest for color recognition, and then the image of the region of interest is affine transformed to enhance the consistency of the input. The image after the projection transformation is converted into the Lab color space, and the image information of six channels in the RGB space is added as the input of the color feature calculation. By collecting bayonet images, a large number of samples after affine transformation are generated, the samples are marked with vehicle colors, and the convolutional neural network is used to learn the marked samples, and the color features of the region of interest are automatically extracted. Using the automatically extracted features and labels The information is learned through the SVM classifier to obtain the final result of color recognition. The color features of the region of interest are calculated and extracted through sample-driven learning, and the obtained features can effectively improve the recognition accuracy of the classifier.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, and relates to a color recognition method of bayonet vehicles. Background technique [0002] At present, road traffic checkpoints, electronic police and other monitoring equipment have been installed and used in large quantities in China. Data mining and target content analysis on the image data collected by existing equipment have become a research hotspot in scientific research and industry. [0003] Vehicle information is the main analysis object in monitoring data such as road traffic checkpoints and electronic police, while the intuitiveness and invariance of rotation and translation of the body color itself have become a very critical information for people to describe and understand vehicle targets. Automatic acquisition of color information and related retrieval services have become a hot demand for criminal investigation and traffic control. [0004] The recognition o...

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): G06K9/20G06K9/62
CPCG06V10/22G06F18/214G06F18/2411
Inventor 尚凌辉高勇郑永宏
Owner ZHEJIANG ICARE VISION TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap