A kind of surveillance video personnel state identification method
A monitoring video and state recognition technology, which is applied in image analysis, image enhancement, instruments, etc., can solve the problems of inaccurate positioning, incapable of self-adaptive adjustment, and inability to realize the discrimination of personnel state
Inactive Publication Date: 2021-04-13
HARBIN INST OF TECH
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] In order to solve the problems of inaccurate positioning existing in the above-mentioned existing algorithms, the inability to self-adaptive adjustment when tracking fails, and the inability to realize the discrimination of the status of personnel in complex indoor scenes, the present invention provides a method for identifying the status of personnel in surveillance video. The technical solution is as follows:
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0054] A monitoring video personnel state recognition method is characterized in that: comprising the following steps:
[0055] (1) Using the target detection algorithm (target detection algorithm based on Faster RCNN convolutional neural network) and target tracking algorithm (mean shift target tracking algorithm) to establish a comprehensive module, the process is as follows Figure 4 As shown, the trajectory information of the human target in the video image is established by using the integrated module established;
[0056] (2) Construct the background model of the video scene based on the time-series gray distribution characteristics of the image pixels, and accurately locate the target boundary information through the background model;
[0057] (3) Judging the state of the human target by the inter-frame position movement of the trajectory.
[0058] The comprehensive module, based on the multi-strategy detection area and tracking area association and allocation method, ...
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
Login to View More Abstract
The invention discloses a monitoring video personnel status identification method. Through the comprehensive module of the existing target detection algorithm and target tracking algorithm, the trajectory information of the human target in the video image is established; the background model of the video scene is constructed based on the gray distribution characteristics of the image pixel time series, and the target boundary information is accurately positioned through the background model; The state of the human target is judged by the inter-frame position movement of the trajectory. The invention solves the disadvantage that a single detection and tracking algorithm cannot judge the status of personnel, and at the same time solves the problems of missed detection and false detection in the detection algorithm and the problems of tracking failure and inaccurate positioning in the tracking algorithm. The results can be given in the indoor monitoring video. There are two states of rest and motion of a person object.
Description
technical field [0001] The invention relates to the field of intelligent processing of video images, in particular to a method for discriminating an object state. Background technique [0002] In view of the indoor monitoring video images obtained by fixed platform, fixed detector field of view and fixed focal length, the scene in the image is complex and there are large changes in posture and shape of the personnel target, relying on manpower to analyze the indoor monitoring video content and identify the The status of people targets is very tedious and costly work. Therefore, it is a very potential development direction to establish an automatic intelligent monitoring system that can accurately automatically extract personnel targets from indoor surveillance videos and determine their status. [0003] At present, the algorithm design for indoor surveillance video images mostly focuses on the detection and tracking of human targets in sequence images. Among them, the dete...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/277G06T7/11G06T7/136G06T7/194G06T7/155
CPCG06T7/20G06T2207/10016G06T2207/20084G06T2207/30241G06T7/11G06T7/136G06T7/155G06T7/194
Inventor 侯晴宇吴鹏赵烨
Owner HARBIN INST OF TECH