Multi-target tracking counting method based on binocular vision

A multi-target tracking and counting method technology, applied in the field of multi-target tracking and counting, can solve the problems of detection equipment that is difficult to accurately distinguish passengers, counting is inaccurate, and passengers get on and off the bus together, so as not to lose the target and improve the counting accuracy , Inhibit the effect of misjudgment

Pending Publication Date: 2022-03-04
深圳市巴视通技术有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Regarding the above-mentioned related technologies, the inventor believes that the detection equipment used in the related technologies needs to detect the passengers one by one when counting the passengers getting on and off the bus. The entry and exit of crowds and multiple targets can be accurately judged, but the detection equipment used in related technologies is difficult to accurately distinguish each passenger, so it is easy to cause inaccurate counting

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
  • Multi-target tracking counting method based on binocular vision
  • Multi-target tracking counting method based on binocular vision
  • Multi-target tracking counting method based on binocular vision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0083] The embodiment of the present application discloses a binocular vision multi-target tracking and counting method.

[0084] refer to figure 1 , a multi-target tracking and counting method for binocular vision, comprising steps S100 to S600,

[0085] Step S100: Simultaneously collect video data from two cameras.

[0086] Step S200: Preprocessing the images in the two video data, and performing stereo matching on the two preprocessed images;

[0087] Among them, preprocessing includes filtering the image in the video. Filtering is the operation of filtering out the frequency of a specific band in the signal, and it is an important measure to suppress and prevent interference;

[0088] After filtering, the normalization of the image is also included. To a certain extent, it can be understood that the pixel value of 0-255 becomes between 0-1, reducing its distribution distance;

[0089] After image normalization, image smoothing is also included. Image smoothing is a kind...

Embodiment 2

[0154] The difference between this embodiment and Embodiment 1 is that: before step S580, steps S573~S574 are also included:

[0155] Match the profile of each passenger's head in the depth map of the adjacent frame;

[0156] Based on the matching results, analyze the similarity of each passenger's head profile in the depth map of adjacent frames, and use the following formula for analysis:

[0157]

[0158] Among them, j represents the j-th passenger head contour in the previous frame image, k represents the k-th passenger head contour in the current frame image, represents the Euclidean distance between the centroids of the two passenger head contours, and represents the The similarity of the gray scale of the contour of the body represents the similarity of the contour area of ​​the head of two passengers.

[0159] Combined with the actual situation of the bus, factors such as bus steps, noise, depth map deviation, etc., make the feature quantity of the same passenger u...

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 the technical field of image recognition, in particular to a multi-target tracking counting method based on binocular vision. Video data in the two cameras are collected at the same time; preprocessing the images in the two pieces of video data, and performing stereo matching on the two preprocessed images; obtaining a depth map, processing the depth map, and extracting a passenger head contour in the depth map based on the processed depth map; performing region duplicate removal on the depth map; tracking passengers in the depth map; and finally calculating the number of passengers based on a tracking result. Binocular vision is adopted to count passengers getting on and off a bus, multiple passengers can be tracked at the same time, the tracking effect is stable, targets are not prone to being lost, meanwhile, images of the passengers close to each other can be segmented and smoothed, each passenger in a depth map can be accurately recognized, and the accuracy of the passengers is improved. And therefore, the counting accuracy of the public transport passenger flow can be improved.

Description

technical field [0001] The present application relates to the technical field of image recognition, in particular to a binocular vision multi-target tracking and counting method. Background technique [0002] Image recognition refers to the technology of using computers to process, analyze and understand images to identify targets and objects in various patterns. In recent years, with the continuous development and innovation of science and technology, the related video algorithms in the field of image recognition are increasingly Perfect, image recognition is widely used in many fields of human life. Research on people counting in public places based on video is one of the important directions in the field of image recognition. Doing a good job of counting people has potential value in both business and life. At this stage, research on people counting in public places based on video It is mainly used in public places such as buses, subway stations and shopping malls. [0...

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): G06T7/246G06T7/62G06T7/13G06T7/12
CPCG06T7/246G06T7/62G06T7/13G06T7/12G06T2207/30241G06T2207/30196G06T2207/20152G06T2207/30242
Inventor 吴国才
Owner 深圳市巴视通技术有限公司
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