Unlock instant, AI-driven research and patent intelligence for your innovation.

Abnormal behavior identification method and system based on skeleton extraction

A skeleton extraction and recognition method technology, applied in neural learning methods, character and pattern recognition, neural architecture, etc., can solve problems such as inability to identify construction site workers and achieve accurate detection results

Inactive Publication Date: 2021-11-23
JIANGNAN UNIV
View PDF8 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] In view of the above analysis, the present invention aims to provide a method for identifying abnormal behaviors based on skeleton extraction to solve the problem that the prior art cannot accurately identify abnormal behaviors of construction site workers

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
  • Abnormal behavior identification method and system based on skeleton extraction
  • Abnormal behavior identification method and system based on skeleton extraction
  • Abnormal behavior identification method and system based on skeleton extraction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0069] This embodiment provides a method for identifying abnormal behavior based on skeleton extraction. The flow chart is as follows figure 1 shown, including:

[0070] Step S1: Obtain a video with a human body image;

[0071] A wide-angle camera with a resolution of 720P is used to shoot the construction area at an angle of obliquely upward and downward. The optical axis of the camera is parallel to the inspection robot. The shooting angle covers the entire work site and collects surveillance video images with a slight overlooking angle.

[0072] Step S2: performing target person detection and tracking on the human body image in the video;

[0073] This embodiment uses the YOLOv3 network to detect human bodies in the video. The YOLOv3 network uses multi-scale fusion for target detection, and has good adaptability to changes in the target scale; the YOLOv3 network uses the K-means algorithm to cluster the initial size of the bounding box. This prior knowledge improves the s...

Embodiment 2

[0127]Based on the same inventive concept, this embodiment provides an abnormal behavior recognition system based on skeleton extraction. Its problem-solving principle is similar to the above-mentioned abnormal behavior recognition method based on skeleton extraction, and the repetition will not be repeated.

[0128] This embodiment provides a skeleton extraction-based abnormal behavior recognition system, including:

[0129] The video acquisition module is used to acquire the video with human body images;

[0130] The detection and tracking module is used to detect the target person in the human body image in the video, and track the detected target person;

[0131] The human body skeleton construction module is used to extract the human body joint confidence map and bone drift field map of the tracked target person, and adopts non-maximum value suppression for the human body joint confidence map to obtain a series of candidate joint points. They are connected to each other ...

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 an abnormal behavior identification method and system based on skeleton extraction. The method comprises the following steps: acquiring a video with a human body image; performing target person detection on the human body image in the video, and tracking the detected target person; extracting a human body joint confidence map and a skeleton drift field map from the tracked target person, and constructing a human body skeleton; and combining human body skeletons of a target person in each frame of image of a video according to a time sequence to obtain a skeleton sequence, constructing a space-time diagram according to the skeleton sequence, carrying out space-time diagram convolution operation on the space-time diagram to carry out behavior feature extraction, classifying the behavior features, and identifying whether the target person has an abnormal behavior or not. According to the method, target person detection and tracking are performed on the video, skeleton extraction is performed, and whether the target person has the abnormal behavior is identified by using the space-time diagram convolutional network, so that accurate detection of the abnormal behavior is realized, and the problem that the abnormal behavior cannot be accurately identified in the prior art is solved.

Description

technical field [0001] The invention relates to the technical field of image processing and behavior recognition, in particular to a method and system for abnormal behavior recognition based on skeleton extraction. Background technique [0002] There are many construction points, wide areas, and heavy tasks in electric power engineering. There are generally high-risk factors such as high-altitude cross-operation, harsh field construction environment, large-scale lifting and hoisting, and personnel electric shock. Insufficient safety management capabilities lead to a high risk of personal accidents. At present, with the continuous increase of the scale of the power grid, the rapid increase of equipment and the acceleration of modern urbanization, the scale and number of construction sites such as power infrastructure, technical transformation, relocation, and overhaul have also increased dramatically. Therefore, power construction sites The demand for security management and...

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): G06K9/00G06K9/46G06T7/246G06K9/62G06N3/04G06N3/08
CPCG06T7/246G06N3/08G06T2207/10016G06T2207/30196G06N3/045G06F18/2415
Inventor 颜文旭吴晨樊启高
Owner JIANGNAN UNIV