Adaptive characteristic block selection-based gait identification method

A technology of gait recognition and feature selection, which is applied in the field of image processing, can solve the problems of not completely avoiding, affecting the effectiveness of gait recognition, and consuming calculations, so as to achieve the effect of rapid recognition

Active Publication Date: 2011-07-13
上海人工智能研究院有限公司
View PDF3 Cites 24 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method does not completely avoid the interference of clothing and objects on the area above the legs of the binary image of the human body, which affects the effectiveness of gait recognition to a certain extent.
In addition, this method makes pixel value change statistics for all pixels of multiple energy maps, which consumes a certain amount of calculation, and there is still room for improvement in its calculation 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
  • Adaptive characteristic block selection-based gait identification method
  • Adaptive characteristic block selection-based gait identification method
  • Adaptive characteristic block selection-based gait identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0022] Such as figure 2 and Figure 7 As shown, this embodiment includes the following steps:

[0023] The first step, using the Least Median of Squares method (Least Median of Squares) to restore the background image B from the N-frame image sequence I of the video (x，y) , and then calculate the current image frame I t and background image B (x，y) The extraction function value of , the foreground image I' is separated by thresholding the function value t (x, y).

[0024] The background image Where: (x, y) is the pixel position, median(·) represents the median filter, t represents the sequence number of the image frame, t={1, 2, ..., N}, min(·) represents the minimization operator.

[0025] Described extraction function value refers to:

[0026] f ( I t ( x ...

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 adaptive characteristic block selection-based gait identification method in the technical field of image processing. By blocking a human body region in a gait energy diagram, acquiring statistical information of each region according to the training data of no clothes and carried object state interference and calculating the statistical information through blocks duringtesting to automatically select characteristic blocks in accordance with the statistical information laws of the training data, the adverse effects of pedestrian clothes and carried object state change on gait identification are effectively overcome, meanwhile, the characteristics of the parts such as head, shoulders and the like of a person are kept, and the effectiveness of the gait identification is improved. Any priori knowledge is not needed, the calculation complexity of characteristic selection is reduced by 28 percent compared with the conventional gait identification method for non-supervision characteristic selection, and meanwhile, the average identification rate is comparable in the conventional supervision characteristic selection-based gait identification method.

Description

technical field [0001] The invention relates to a method in the technical field of image processing, in particular to a gait recognition method based on adaptive feature block selection. Background technique [0002] As the only remotely identifiable biometric feature, gait has important applications in video surveillance. Human gait is not easy to camouflage and hide, and does not require the cooperation of the observer during the monitoring process, which provides a reliable basis for gait recognition. Most of the existing gait recognition methods adopt similar steps. Usually, the gait video sequence of an object is first separated from the foreground and background; then a certain gait feature is extracted from the human body pixel block in the foreground area, and a gait classifier is constructed by a supervised learning method; When performing identity recognition, the same gait features are extracted from the current video sequence and sent to the classifier to obtai...

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): G06K9/62G06T7/00
Inventor 徐奕杨小康李宁
Owner 上海人工智能研究院有限公司
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