Video-based target long-term tracking method

A long-term tracking and target tracking technology, which is applied in the field of long-term target tracking based on video, can solve problems such as complex background environment interference, tracking algorithm tracking failure, etc., to meet real-time requirements, reduce the probability of tracking errors, improve accuracy and The effect of robustness

Inactive Publication Date: 2016-12-07
屈桢深
View PDF5 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Although video target tracking technology has made great progress, and some results have entered the practical stage, there are still many problems: (1) Tracking targets are diverse
(4) Complex background environment interference
The tracking algorithm is easily disturbed by the background information, which leads to tracking failure

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
  • Video-based target long-term tracking method
  • Video-based target long-term tracking method
  • Video-based target long-term tracking method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] to combine figure 1 , 2 , 3 illustrates the patent of the present invention, a kind of long-term target tracking method based on video described in the patent of the present invention, comprises the following steps:

[0033] (1), read the video, read in the first frame of the video, and obtain the target tracking frame by selecting or reading in the position parameters of the target frame with the mouse;

[0034] Read the video, read in the first frame of the video, print out the first frame of the video in the image window on the screen, select the target frame in the image window through the mouse or read in the parameters of the target frame through text, so as to get the initial Parameter information of the target box.

[0035] (2), the system is initialized, the scanning window is set up, and the variance classifier, set classifier and nearest neighbor classifier in the training detector;

[0036] Based on the size of the initial target frame, the scanning windo...

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 video-based target long-term tracking method. The method comprises the steps of reading a first frame of a video, and obtaining a target tracking frame; establishing scanning windows, and training three classifiers in a detector; reading a next frame of image from the video; predicting the size and position of a target tracking frame in a current frame of image by a tracker through the target tracking frame in the previous frame; entering the detector, limiting a detection area through Kalman filtering and direction detection, reducing the number of scanning windows needed to be detected, enabling the scanning windows to enter the three cascaded classifiers in sequence, dividing the scanning windows into two parts by utilizing a multi-thread processing technology, and putting the two parts into two threads respectively for detecting image blocks in the scanning windows, thereby obtaining all target-containing image blocks; obtaining a final output by a synthesizer according to output results of the tracker and the detector, and judging whether the detector is updated or not; and updating the detector, and printing out a finally output target tracking frame on a screen.

Description

technical field [0001] The invention belongs to the field of video target tracking, and more specifically relates to a video-based long-term target tracking method. Background technique [0002] Video tracking technology refers to the process in which the computer continuously infers the state of the target in the video sequence. Its task is to generate the trajectory of the target by locating the target in each frame of the video, and provide a complete target area at each moment. There are many types of target tracking algorithms, and they can generally achieve the desired effect, but they lack versatility. The more commonly used target tracking algorithms are as follows: tracking algorithms based on target motion characteristics, such as frame difference method, optical flow-based tracking methods, etc.; tracking algorithms based on target feature parameters, such as contour-based tracking algorithms, feature-based tracking algorithms, etc. Point tracking algorithms, etc...

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): G06T7/20G06K9/62
CPCG06T2207/10016G06T2207/20081G06T2207/20021G06F18/24
Inventor 吕枭梁亮徐铎马婉玉
Owner 屈桢深
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