Feature extraction and recognition method of vehicle logo based on feature quantization based on gradient direction division

A gradient direction and feature extraction technology, applied in the field of target recognition, to achieve high recognition rate, increase feature description information, and reduce dimensionality.

Active Publication Date: 2022-03-04
INTELLIGENT MFG INST OF HFUT
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In actual situations, due to factors such as illumination, size, and orientation, many bayonet images are not idealized car logo images

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
  • Feature extraction and recognition method of vehicle logo based on feature quantization based on gradient direction division
  • Feature extraction and recognition method of vehicle logo based on feature quantization based on gradient direction division

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] Such as figure 1 As shown, a feature extraction method based on gradient direction division includes the following steps:

[0038] (1) Extract the gradient size and gradient direction information of the car logo image:

[0039] The collected car logo samples are divided into two categories, training samples and test samples. Normalize and grayscale the training samples, calculate the gradient size Gv and gradient direction Go of each pixel in the sample image, and store the gradient size and gradient direction information in the gradient matrix.

[0040] (2) Gradient direction division and gradient size matrix generation:

[0041] The gradient direction is divided into k ranges, and for each pixel, the pixels belonging to the same range in its neighborhood are stored in a matrix (called a gradient size matrix). Then each car logo sample has k gradient size matrices, and then calculate the gradient size matrix of each car logo sample.

[0042] (3) Feature extraction ...

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 discloses a vehicle logo feature extraction and recognition method based on feature quantization of gradient direction division. Firstly, the vehicle logo image taken by the bayonet system is preprocessed, the gradient size and direction of each pixel are calculated, and all pixels Gradient information is stored in the corresponding gradient matrix; k gradient directions are pre-divided, and the gradient sizes of all pixels around each pixel corresponding to k gradient directions are counted, and the gradient sizes are accumulated into k different gradient size matrices; k are extracted respectively For the LTP feature of the gradient size matrix, the extracted k LTP features are spliced ​​to obtain the pixel features of the original car logo image; all the features in the sample are classified by K-means to obtain an offline codebook, and finally the SVM is used for the car logo Classification and recognition of images. The invention proposes a specific recognition scheme for the vehicle logo recognition in the bayonet image, and the recognition result has a high accuracy rate, which can meet the requirements of the actual intelligent transportation system.

Description

technical field [0001] The invention relates to the technical field of target recognition, in particular to a method for extracting and recognizing vehicle logo features based on feature quantization of gradient direction division. Background technique [0002] In the application of computer vision, vehicle logo recognition has always been an important part of the intelligent transportation system. It is widely used in the fields of collecting vehicle information, analyzing traffic flow, identifying vehicles that violate regulations and causing accidents, and regulating traffic order. In the intelligent transportation system, the high-definition bayonet system uses advanced photoelectric technology, image processing technology, and pattern recognition technology to take pictures of every passing car. The vehicle logo feature extraction and recognition method in the present invention is aimed at bayonet high-definition images, and uses the existing car logo positioning techno...

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 & AuthorityPatents(China)
IPC IPC(8): G06V20/54G06V10/46G06V10/762G06V10/764
CPCG06V20/584G06V10/473G06V10/462G06F18/23213G06F18/214
Inventor余烨徐京涛路强薛峰
OwnerINTELLIGENT MFG INST OF HFUT