High-density pedestrian detection method

A pedestrian detection, high-density technology, applied in the direction of instruments, counting objects, characters and pattern recognition, etc., can solve the problems of complex calculation, low detection accuracy and robustness, poor timeliness, etc., to achieve small amount of calculation and strong versatility , the effect of video data reduction

Inactive Publication Date: 2014-03-19
NORTH CHINA UNIVERSITY OF TECHNOLOGY
View PDF4 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] For this reason, the technical problem to be solved by the present invention is that the detection accuracy and robustness of the high-density pedestrian detection method are low, the calculat...

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
  • High-density pedestrian detection method
  • High-density pedestrian detection method
  • High-density pedestrian detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] This embodiment provides a high-density pedestrian detection method, and the flowchart is as follows figure 1 As shown, it includes the following steps: First, a certain amount of video data is selected, and a pedestrian head model is established according to the shape and texture characteristics of the pedestrian's head, as a template for extracting the pedestrian's head information in the video stream from the camera.

[0049] Secondly, for the video stream from the camera, obtain the pedestrian head information in the detection area in each frame of the image according to the pedestrian head model; create a queue for each pedestrian head information in the first frame of image, and The position information of the pedestrian's head information is stored in the queue; each subsequent frame of image determines whether it is the same pedestrian's head information according to the relationship between it and the pedestrian's head information in the previous frame of image; if ...

Embodiment 2

[0053] This embodiment provides a high-density pedestrian detection method, and the flowchart is as follows figure 1 As shown, the specific steps are as follows:

[0054] First, select a certain amount of video data, and build a pedestrian head model based on the shape and texture of the pedestrian's head. The process of selecting a certain amount of video data and establishing a pedestrian head model based on the video data includes: obtaining the head shape and texture characteristics of each pedestrian according to the video data, and then according to the head shape and head shape of each person. Part texture features build the pedestrian head model.

[0055] The process of acquiring the head shape and texture features of each pedestrian according to the video data, wherein the process of acquiring the head shape includes: acquiring the boundary of the pedestrian's head image; extracting the boundary sequence of the pedestrian's head portrait ; Calculate the shape of the pedes...

Embodiment 3

[0070] This embodiment is based on the above-mentioned embodiment. For the video stream from the camera, the process of acquiring the pedestrian head information in the detection area in each frame of the image according to the pedestrian head model also includes: The incoming video stream sets the detection area, the width of the detection area is the width of the effective passage, and the height of the detection area is five times the average diameter of the pedestrian's head.

[0071] In other embodiments, the height of the detection area is four or six to ten times the average diameter of the pedestrian's head, which is specifically selected according to the characteristics of the camera and the traffic scene.

[0072] The high-density pedestrian detection method provided by this embodiment sets an effective detection area for the detected image, greatly reduces the video data that needs to be processed, and has a small amount of calculation, which improves the efficiency of pe...

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

A high-density pedestrian detection method includes the following steps: first building pedestrian head models according to the head shapes and the texture features of pedestrians; processing images of each frame and transmitted by a camera according to the head models, building a queue for information of each pedestrian head in the images of a first frame, and judging whether the information of the pedestrian heads in following images of each frame and the images of a previous frame is the information of the same pedestrian head; on yes judgment, the information is stored into a corresponding queue, on no judgment, generating a new queue; sequentially processing the following images of each frame according to the method. A pedestrian is detected if a queue with the position information updated continuously and containing information of at least three heads is generated. The pedestrian head models are acquired through the head shapes and the texture features of the pedestrians, the information of the pedestrian head in the images of each frame is processed, the method has the advantages of being high in detection accuracy, good in robustness, convenient to compute, high in timeliness and the like, and the method has wide application prospects in the aspects of intelligent traffic video monitoring, robot vision and the like.

Description

Technical field [0001] The invention relates to a high-density pedestrian detection method. Specifically, it is a high-density pedestrian detection method suitable for high-density scenes of large pedestrian traffic, and belongs to the technical field of urban traffic management. Background technique [0002] There are many advanced pedestrian statistics systems at home and abroad, but they are all aimed at the situation of sparse pedestrians, and they can't do anything about crowded areas such as airports and stations. Detecting and tracking pedestrians in high-density people is a difficult problem. Common methods such as detecting motion areas through background models and performing foreground segmentation, or detecting human bodies through human body models and features, cannot solve the problem of pedestrian detection in high-density crowds. The reason is: in high-density crowd videos, the observation area Most or all of them are in motion, and there is serious obstruction...

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): G06K9/62G06M11/00
Inventor 张师林熊昌镇刘小明
Owner NORTH CHINA UNIVERSITY OF TECHNOLOGY
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