Self-adaptive crowd grouping detection method
A cluster detection and adaptive technology, applied in the field of computer vision and intelligent transportation, can solve problems such as over-classification, and achieve the effect of reducing clustering errors, wide application prospects, and good effects.
Pending Publication Date: 2019-07-05
SHANGHAI UNIVERSITY OF ELECTRIC POWER
View PDF3 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
This multi-layer clustering detection method does not require prior information, but this method is prone to over-classification
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
[0046] Such as figure 1 Shown is the overall schematic flow chart of detection method of the present invention, comprises the following steps:
[0047] Step 1: Establish a background model for each frame of the video image and obtain a black and white image;
[0048] Step 2: extract feature points in the original video image corresponding to the white area in the black and white image, and obtain the coordinates of each frame image feature point after inter-frame tracking;
[0049] Step 3: use the distance information between feature points to perform the first clustering process;
[0050] Step 4: Use the acceleration direction information between the feature points to perform the second clustering process on the result of the first clustering process and form a multi-layer cluster;
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 relates to a self-adaptive crowd grouping detection method. The method comprises: KLT feature points are extracted from a foreground area obtained through a background removal method ofthe Gaussian mixture model, the motion characteristics of the feature points are analyzed, the distance between the feature points and the acceleration direction are used as input of hierarchical clustering, and the feature points in all the foregrounds are traversed to achieve clustering detection; and then, analyzing social force borne by the clustering centers in the layering result to obtain angles of resultant force directions among the clustering centers so as to judge whether the clustering centers need to be merged or not. Compared with the prior art, unsupervised automatic crowd grouping can be realized for disordered motion intensive scenes in public places, and no priori knowledge is needed. The method has wide application prospects in the aspects of target detection, intelligent video monitoring, intelligent traffic planning and the like.
Description
technical field [0001] The present invention relates to the technical fields of computer vision and intelligent transportation, in particular to an adaptive grouping detection method for crowds. The method uses a multi-layer clustering filter model to group and detect crowds in dense scenes to realize crowd movement in irregular dense motion scenes. Classification of features and grouping of people. It is widely used in target detection and intelligent video surveillance systems. Background technique [0002] In recent years, with the rapid development of the country's economy, some emerging public places for people's life and entertainment have emerged as the times require, and the frequency of group sexual behavior has also greatly increased, and the harm to society has increased. Mass incidents and the protection of social security promote the development of intelligent monitoring. As a hot field intersecting multiple fields, intelligent video surveillance has always be...
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 Applications(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V20/53G06F18/2321G06F18/241
Inventor 赵倩程祥
Owner SHANGHAI UNIVERSITY OF ELECTRIC POWER