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

Multiple target tracking-based passenger flow statistics method

A multi-target tracking and statistical method technology, which is applied in the field of video surveillance and pedestrian target counting, can solve the problems of poor adaptability, inability to deal with human body occlusion, and inability to adapt to people flow statistics, etc.

Inactive Publication Date: 2017-04-26
ZHENGZHOU JINHUI COMP SYST ENG
View PDF11 Cites 35 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the recognition and tracking method requires high video quality, and the camera needs to shoot the human body with a vertical perspective (TOP-VIEW). The camera angle cannot be adjusted, which leads to poor adaptability, cannot deal with human body occlusion, and cannot adapt to the traffic flow at different angles. Statistics; the existing equipment cannot be used, and special front-end acquisition equipment and other supporting hardware need to be reinstalled, which is costly and cannot protect the customer's existing equipment investment; due to some inherent characteristics of the human body and the complexity of the actual scene, such as pedestrians and pedestrians The overlapping occlusions, different postures, different clothing, background differences and brightness differences between people, the existing people counting technology can not achieve good results

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
  • Multiple target tracking-based passenger flow statistics method
  • Multiple target tracking-based passenger flow statistics method
  • Multiple target tracking-based passenger flow statistics method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] Embodiment one, see figure 2 As shown, a passenger flow counting method based on multi-target tracking, specifically includes the following steps:

[0050] Step 1. Use the deformable part model to carry out offline training on the marked humanoid sample to obtain the humanoid part model. The humanoid part model includes a root filter, a part filter, and a spatial model of the part relative to the root position;

[0051] Step 2. Use the humanoid part model to perform humanoid target detection on each frame of the video;

[0052] Step 3. Use the two-stage trajectory association method to track the pedestrian target in the image scene to obtain the motion trajectory of the detected pedestrian;

[0053] Step 4. In the visible area of ​​the camera, determine the moving direction of the pedestrian target according to the moving track, and count the pedestrian target in real time according to the moving track and moving direction.

Embodiment 2

[0054] Embodiment two, see Figure 2-5 As shown, a passenger flow statistics method based on multi-target tracking, the specific implementation is as follows:

[0055] 1) Annotate the humanoid sample, mark the position of the humanoid in the image and the position of each component in the image, use the DPM algorithm to perform offline training on the humanoid sample, and obtain the humanoid part model, which includes the root filter and the component filter , and the spatial model of the part relative to the root position.

[0056] 2) Use the humanoid component model to perform humanoid target detection on each frame of the video. The specific content is as follows: use image smoothing and downsampling to generate an image pyramid, calculate HOG features for each layer of the image pyramid, and obtain the feature map of the layer image ;For each feature map of the pyramid, use a fixed-size sliding window to slide, and calculate the score of each sliding window, where the siz...

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 relates to a multiple target tracking-based passenger flow statistics method which comprises the following steps: a deformable part model is used for subjecting a marked human shape sample to off line training operation, and a human shape part model can be obtained; the human shape part model comprises a root filter, a part filter and a space model for positions of parts relative to a root; the human shape part model is used for subjecting each image frame in a video to human shape object detection operation, a two stage track association method is adopted for tracking pedestrian objects in image scenes, movement tracks of pedestrians being detected are obtained, and the pedestrian objects can be counted in real time according to the movement tracks and movement directions in a camera-covered zone. Front end cameras adopted in the multiple target tracking-based passenger flow statistics method are deployed above entries and exits of indoor environments such as supermarkets, supermarkets and the like; vertical or overlooking sampling operation can be performed; tilting sampling scenes can be optimized in a specific manner, the objects can be detected in the camera-covered zone, the number of people going in and out can be counted in real time while the pedestrians are detected and tracked, and correct rates of pedestrian object statistics exceed 90%.

Description

technical field [0001] The invention relates to the field of counting and counting pedestrian targets in video surveillance, in particular to a method for counting passenger flow based on multi-target tracking. Background technique [0002] The main purpose of people flow statistics is to make accurate and real-time statistics on the number of pedestrians in the video surveillance system. It is mainly used in various places of activity, such as supermarkets, restaurants, shopping malls, etc. By grasping the information of the number of people in real time, it is possible to rationally dispatch manpower and material resources to improve service quality. Therefore, people counting technology is of great significance in the field of intelligent video surveillance and computer vision, and it is a research topic with practical application value and urgent development. [0003] The applicability of a people counting system refers to whether the system can operate normally under ...

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 Applications(China)
IPC IPC(8): G06T7/292G06K9/00
CPCG06T2207/20081G06T2207/10016G06T2207/30242G06T2207/30232G06V40/10G06V20/44G06V20/40
Inventor 张晨民彭天强李丙涛栗芳
Owner ZHENGZHOU JINHUI COMP SYST ENG
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