Improved gradient histogram feature extraction algorithm based on texture trend and its application method

A gradient histogram and feature extraction technology, applied in computing, computer components, instruments, etc., can solve problems such as classification accuracy decline, image illumination, scale and rotation sensitivity, etc., to achieve elimination of influence, fast speed, and high classification accuracy Effect

Inactive Publication Date: 2019-03-22
XIAN UNIV OF POSTS & TELECOMM
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the deficiencies in the prior art, the purpose of the present invention is to provide an improved gradient histogram feature extraction algorithm and its application method based on the texture trend, which mainly solves the problems of the existing tire pattern database classific

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
  • Improved gradient histogram feature extraction algorithm based on texture trend and its application method
  • Improved gradient histogram feature extraction algorithm based on texture trend and its application method
  • Improved gradient histogram feature extraction algorithm based on texture trend and its application method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] Example 1: In order to verify the classification performance of this method, the performance of HOG-TT is compared with other existing techniques. To this end, different image features are used to train the SVM classifier, and various SVM parameters such as the kernel function of each feature are also adjusted to the optimal parameters. The features used include the HOG-TT feature of this method, the texture feature based on discrete wavelet transform (DiscreteWavelet Transform, DWT), the texture feature based on curvelet domain energy distribution (Curvelet Domain Energy Distribution Algorithm, CEDA), and the color histogram based on HSV The color feature (HSV color histogram, HSV), the texture feature based on the equivalent rotation invariant local binary pattern (Rotation Invariant Uniform Local Binary Pattern, Riu-LBP), the HOG feature, the texture feature based on the compressed gradient histogram (Compressed Histogram of Gradients, CHoG), texture features based o...

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 an improved gradient histogram feature extraction algorithm based on texture trend and an application method thereof. Comprises the following five modules: (1) a gradient calculation module for extracting pixel point gradient amplitude and gradient direction; (2) a statistical module defining circular statistical units and performing directional gradient histogram statistics on each unit; 3) alignment module that extracts that texture trend, calculating the trend direction, and align the alignment module of the eigenvector of the unit according to the trend direction; (4) a construction model that connects unit eigenvector to construct HOG-TT feature; (5) a clasisification training module Using the trained SVM model group to analyze the input HOG-TT feature classification or HOG using training data-TT feature training SVM. The invention adds excellent rotation invariance on the premise of keeping HOG characteristic illumination and scale invariance, and reducesthe dimension of the characteristic according to the characteristic of the tire pattern image, and reduces the computational complexity.

Description

technical field [0001] The invention relates to the technical field of image classification in digital image processing, in particular to an improved gradient histogram feature extraction algorithm based on texture trends and a method for classifying tire pattern images. Background technique [0002] The texture feature reflects the change of the gray value of the image. It is one of the important underlying features of the image. It is the own characteristic of the object material itself and does not change with the outside world. Therefore, extracting excellent texture features can overcome the influence of external environment changes on image features, thereby improving the classification performance of the method. The Histogram of Oriented Gradient (HOG) feature is a feature that is insensitive to external illumination and scale changes. [0003] The HOG feature is a texture feature with illumination and scale invariance, and is often used in pedestrian detection algor...

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
IPC IPC(8): G06K9/46G06K9/62
CPCG06V10/50G06F18/2411
Inventor 刘颖葛瑜祥张帅王富平
Owner XIAN UNIV OF POSTS & TELECOMM
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