Automatic recognition method of vehicle logo based on principal component analysis convolutional neural network

A convolutional neural network and principal component analysis technology, applied in biological neural network models, neural architecture, character and pattern recognition, etc., can solve the problem of recognition rate and recognition speed need to be improved, achieve high classification accuracy, improve processing Efficiency, effect of simplified convolution process and training method

Active Publication Date: 2018-08-28
JIANGSU AEROSPACE DAWEI TECH CO LTD
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, the existing technologies of vehicle identification include the classification method using feature invariant moment distance, the recognition method based on SIFT features, etc., which need to be improved in recognition rate and recognition speed

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
  • Automatic recognition method of vehicle logo based on principal component analysis convolutional neural network
  • Automatic recognition method of vehicle logo based on principal component analysis convolutional neural network
  • Automatic recognition method of vehicle logo based on principal component analysis convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] A method for automatic recognition of car logos based on principal component analysis convolutional neural network, including ideal output feature vector T for classification of various car logos k Obtaining steps and vehicle logo recognition steps, wherein, T k Indicates the ideal output feature vector of car logo classification, k represents the number of car logo types,

[0044] The ideal output feature vector T of the classification of various vehicle logos k The steps taken include:

[0045] Collect N copies of various car logo images as sample images. In this embodiment, N can be 5000 or 6000, and the various car logo images are respectively positioned to obtain various car logos with a size of 44×44 pixels N parts of the grayscale image are accurately positioned in the region, and then through training the convolutional neural network, determine and obtain the ideal output feature vector T of various vehicle logo classifications. The training method of the 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

A method for automatic recognition of car logos based on principal component analysis convolutional neural network, including the steps of obtaining ideal output feature vectors of various car logo classifications and the steps of car logo recognition, wherein, represents the ideal output feature vectors of car logo classifications, and k represents car logo classification. Mark kind number, described all kinds of car mark classification ideal output feature vector is to use N parts of all kinds of car mark image samples to convolutional neural network to train and obtain, and described car mark recognition is to obtain the output vector of car mark to be identified After Z, by calculating the Euclidean distance and degree of belonging between the ideal output feature vector of the car logo classification of each brand and the output vector Z of the corresponding car logo to be recognized, the corresponding brand car logo in the maximum degree of attribution is the car logo to be recognized .

Description

technical field [0001] The technical field of vehicle feature detection in traffic images, in particular, relates to an automatic identification method for vehicle logos based on principal component analysis convolutional neural network. Background technique [0002] As an important part of the technical field of vehicle feature detection in traffic images, vehicle logo recognition can obtain vehicle information more accurately, and has been more and more widely used in the automatic recording of vehicle whereabouts and illegal vehicles. At present, the existing technologies for vehicle logo recognition include the classification method using feature invariant moment distance, SIFT-based feature recognition method, etc., which need to be improved in recognition rate and recognition speed. [0003] Convolutional neural network (CNN) is a kind of artificial neural network, which is mainly used to recognize two-dimensional graphics with displacement, scaling and other forms of ...

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): G06K9/62G06K9/32G06N3/04
CPCG06V10/25G06V2201/08G06N3/045G06F18/24
Inventor 狄明珠韩晶方亚隽
Owner JIANGSU AEROSPACE DAWEI TECH CO LTD
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