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

Tire trace image feature extraction method in local gradient direction three-valued mode

An image feature extraction and local gradient technology, applied in the field of image processing, can solve the problems of complex edge texture lines, single trace color, etc., and achieve the effect of high retrieval accuracy, high average precision, and improved retrieval accuracy.

Active Publication Date: 2019-11-22
XIAN UNIV OF POSTS & TELECOMM
View PDF9 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At the same time, the edge grain lines are very complicated, and the edge grain lines of different types of tires have different widths and directions, and the grain lines are interconnected and entwined, and the color of the traces is relatively single

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
  • Tire trace image feature extraction method in local gradient direction three-valued mode
  • Tire trace image feature extraction method in local gradient direction three-valued mode
  • Tire trace image feature extraction method in local gradient direction three-valued mode

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] The image of this embodiment comes from the applicant's self-built tire trace image database, including 30 categories of 80 images in total, 2400 pieces. Experiments have been carried out. The steps of the tire trace image feature extraction method in the local gradient direction ternary mode are as follows (see figure 1 ):

[0043] (1) Image preprocessing

[0044] From the tire trace image database, a total of 2,400 tire trace sample images of 30 categories and 80 images in each category were selected, and the size was normalized to 384×384 and gray-scaled.

[0045] (2) Feature extraction

[0046] 1) Use the Sobel edge detection method to determine the image difference G of the image along the x directionx and the image difference G in the y direction y , use the formula (1) to determine the gradient direction angle α(x,y) of each pixel in the image

[0047] α(x,y)=arctan(G y / G x ) (7)

[0048] 2) In the 3×3 neighborhood sliding window, use the local gradient di...

Embodiment 2

[0073] The image of this embodiment comes from the tire trace image database built by the applicant, including 30 categories of 80 images in total, 2400 pieces. Experiments are carried out. The steps of the tire trace image feature extraction method in the local gradient direction ternary mode are as follows:

[0074] (1) Image preprocessing

[0075] This step is the same as in Example 1.

[0076] (2) Feature extraction

[0077] 1) Use the Sobel edge detection method to determine the image difference G of the image along the x direction x and the image difference G in the y direction y , use the formula (1) to determine the gradient direction angle α(x,y) of each pixel in the image

[0078] α(x,y)=arctan(G y / G x ) (13)

[0079] 2) In the 3×3 neighborhood sliding window, use the local gradient direction ternary mode LGDTP method to determine each gradient direction angle value, and the local gradient direction ternary mode LGDTP method adds custom thresholds t, g i grea...

Embodiment 3

[0085] The image of this embodiment comes from the tire trace image database built by the applicant, including 30 categories of 80 images in total, 2400 pieces. Experiments are carried out. The steps of the tire trace image feature extraction method in the local gradient direction ternary mode are as follows:

[0086] (1) Image preprocessing

[0087] This step is the same as in Example 1.

[0088] (2) Feature extraction

[0089] 1) Use the Sobel edge detection method to determine the image difference G of the image along the x direction x and the image difference G in the y direction y , use the formula (1) to determine the gradient direction angle α(x,y) of each pixel in the image

[0090] α(x,y)=arctan(G y / G x ) (16)

[0091] 2) In the 3×3 neighborhood sliding window, use the local gradient direction ternary mode LGDTP method to determine each gradient direction angle value, and the local gradient direction ternary mode LGDTP method adds custom thresholds t, g i grea...

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 tire trace image feature extraction method in a local gradient direction three-valued mode. The tire trace image feature extraction method comprises the steps of image preprocessing, feature extraction, feature vector determination and image retrieval. The invention provides a local gradient direction three-valued mode feature suitable for a tire trace image. A more stable gradient direction value is adopted to replace a gray value to carry out local texture coding. The threshold quantification is carried out on a central pixel gradient direction angle, high-quality texture edge information is generated. The retrieval accuracy is improved. The similarity calculation is carried out on the features described by the feature vectors by using a Manhattan distance to obtain a retrieval result. The retrieval accuracy is obviously superior to that of other texture features. The method has the advantages of being clear in tire trace image texture edge information, highin retrieval accuracy, high in average precision ratio, suitable for large sample data and the like, and can be used for tire trace image feature extraction.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to an image feature extraction method. Background technique [0002] Usually in a traffic accident, the relationship between the tire indentation traces left on the scene and the physical evidence is used to explain the accident process and judge the responsibility of both parties, and the tire traces on the ground are often one of the most useful traces of physical evidence, so tire trace retrieval is often used in public security. Obtaining clues in solving crimes or dealing with traffic accidents. [0003] The research on tire trace image retrieval in my country started late, and the research on tire traces is even less fruitful, and there is no standard tire trace image test database in the field of tire trace image retrieval and classification research. At present, the traditional method of feature combination is mainly used to classify and retrieve tire t...

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): G06F16/583G06T5/40G06T7/13
CPCG06F16/5862G06T7/13G06T5/40G06T2207/30108
Inventor 刘颖董海涛
Owner XIAN UNIV OF POSTS & TELECOMM
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