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

A Method of Image Feature Extraction

An image feature extraction and feature value technology, applied in the field of image processing, can solve problems such as poor real-time performance, slow speed, and hysteresis, and achieve the effect of reducing the amount of calculation and improving the overall speed

Active Publication Date: 2022-03-15
FORYOU GENERAL ELECTRONICS
View PDF13 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In the specific calculation process, the cells move on the block, and the block is also moving, so there are actually many repeated cell operations, and there are a large number of cells in the entire block. According to this method, the 128*64 image, It takes about 0.14 seconds to run once on a dual-core desktop computer. For advanced driver assistance system (ADAS) products, the image resolution is generally 1280*720. Therefore, although the accuracy of this method is high, the calculation is complicated and the feature generation process is lengthy. , resulting in slow speed, poor real-time performance, and a great sense of hysteresis

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 Method of Image Feature Extraction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] The implementation of the present invention will be described in detail below in conjunction with the accompanying drawings. The accompanying drawings are only for reference and description, and do not constitute a limitation to the protection scope of the present invention.

[0050] Embodiments of the present invention provide a method for extracting image features, including:

[0051] Step 1. Set the size of the gradient operation matrix.

[0052] In this embodiment, a 7*7 gradient operation matrix is ​​used, and the X and Y gradients of the pixel (x, y) are the grayscale operation results of itself, the first 3 pixels, and the last 3 pixels. That is to say, the gradient obtained in this way includes the pixel information of the front and back 3 rows and the front and back 3 columns.

[0053] Step 2. Calculate the gradient of each pixel of the gradient operation matrix.

[0054] Specifically include the following steps:

[0055] Step 21. Calculate the initial grays...

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 present invention provides an image feature extraction method, the method comprising: the step of setting the size of the gradient operation matrix; the step of calculating the gradient of each pixel of the gradient operation matrix; the step of determining the gradient interval to which the gradient of each pixel belongs; according to the step of each pixel The step of calculating the gradient length of the gradient of ; the step of calculating the feature value of the block. The invention realizes a substantial increase in image feature extraction speed.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an image feature extraction method. Background technique [0002] HOG (Histogram of Oriented Gradient, histogram of gradient orientation) is an image descriptor proposed by Dalal et al. of the French National Institute of Computer Science and Automatic Control at the 2005 CVPR conference to solve human target detection. This method uses gradient Direction histogram features are used to express the human body, and the shape information and motion information of the human body are extracted to form a rich feature set. [0003] The process of HOG feature extraction is: divide an image into one or more detection windows (WIN), and the general detection window size is 128*64 by default. The detection window is composed of multiple blocks (BLOCK), each block is composed of 16 adjacent pixels, and each block is divided into several (for example, 4) cells (CELL) of the same siz...

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): G06V10/50
CPCG06V10/50
Inventor 侯太炎徐焕东张春裕
Owner FORYOU GENERAL ELECTRONICS
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