Method for pedestrian matching between viewpoints of non-overlapped cameras

A matching method and camera technology, which can be used in closed-circuit television systems, computer components, instruments, etc., can solve the problems that feature points are difficult to extract and affect the matching effect, etc.

Active Publication Date: 2011-11-09
SUN YAT SEN UNIV
View PDF3 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method is easily affected by noise. When the pedestrian target is occluded by the outside or the d

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
  • Method for pedestrian matching between viewpoints of non-overlapped cameras
  • Method for pedestrian matching between viewpoints of non-overlapped cameras
  • Method for pedestrian matching between viewpoints of non-overlapped cameras

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] Such as figure 1 As shown, the present invention is a pedestrian matching method between non-overlapping camera viewpoints, comprising the following steps:

[0052] (1) For the set non-overlapping camera C i 、C j , determine C j Target to be matched in O j,b , and camera C i Detected in leaving C for a period of time i Enter camera C j R targets, these targets are extracted separately and saved; for target O j,b , go directly to step (3), for C i Goal in step (2);

[0053] (2) For camera C j A target O detected in j,b , looking for C i Which one of the targets corresponds to it, it needs to be trained according to the corresponding pedestrian targets detected in the two cameras to obtain the cumulative brightness transfer function model between the cameras, that is, the CBTF mapping function, and then apply this function to the camera C i Calibrate all targets in ; go to step (3);

[0054] (3) Carry out the following operations on the target, first calculat...

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 method for pedestrian matching between viewpoints of non-overlapped cameras. The method provided by the invention comprises the following steps of: determining an object Oj,b to be matched in Cj according to the arranged non-overlapped cameras Ci and Cj and detecting R objects which leave the camera Ci and enter the camera Cj within a period of time; carrying out illumination correction on pedestrian objects in the camera Ci and calculating a DLBPriu2 value of each pixel in the objects; then calculating three two-dimensional histograms R_DLBP (Red_DLBP), G_DLBP (Grey_DLBP) and B_DLBP (Black_DLBP) of each object; distributing each two-dimensional histogram into a one-dimensional histogram according to the DLBP of color components; connecting three one-dimensional histograms in series to obtain an RGB-DLBP descriptor; and finally, realizing pedestrian matching among different cameras based on the similarity of the RGB-DLBP descriptor. In the method provided by the invention, color information of a colorful target is combined with space structure information so that the colorful target can be more effectively represented; and simple construction and convenience in calculation are realized, and the requirements of the pedestrian matching between the viewpoints of non-overlapped cameras in the practical application can be met.

Description

technical field [0001] The invention relates to the field of video monitoring, in particular to a pedestrian matching method between non-overlapping camera viewpoints. Background technique [0002] With the improvement of people's safety awareness, video surveillance systems are gradually being used in more and more occasions. Due to the limited field of view of a single camera, and in the process of tracking pedestrians, it is easy to cause tracking loss due to factors such as being blocked, so the current In some large and complex public places, more cameras are used to expand the scope of monitoring and joint tracking. Matching between pedestrians detected by multiple cameras, that is, pedestrian matching between non-overlapping camera viewpoints, has become a research direction that has attracted much attention in video surveillance research. In the monitoring process, human motion is the focus of our monitoring, but since human motion is a non-rigid body motion, its tr...

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): G06K9/00H04N7/18
Inventor 赖剑煌连国云郑伟诗
Owner SUN YAT SEN UNIV
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