A method and system for extracting human behavior features for edge computing
A feature extraction and behavior technology, applied in the field of human behavior recognition, can solve the problems of increased data volume, large bandwidth, and restrictions on large-scale applications, and achieve the effect of reducing network bandwidth occupation and low latency requirements
Active Publication Date: 2022-05-20
TSINGHUA UNIV
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] In related technologies, the video data collected by the visual sensor needs to be uploaded to the cloud server for centralized processing. On the one hand, as the number of visual sensors increases, the amount of data uploaded to the cloud server increases significantly, occupying a large amount of bandwidth; on the other hand, human behavior The recognition tasks are all completed by the cloud server, and the computing resources cannot be dynamically scheduled, resulting in high computing pressure on the cloud server, which limits the large-scale application of human behavior recognition methods in the security field
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 Construction
[0066] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
[0067] refer to figure 1 , the present invention proposes an edge computing-oriented human behavior feature extraction system, the system comprising:
[0068] a visual sensor, the visual sensor is used to collect video data, and transmit the video data to the first edge node connected thereto;
[0069] The first edge node, the first edge node is used to receive the video data transmitted by the visual sensor, calculate the spatial position of the key point ac...
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 present application provides a method and system for extracting human behavior features oriented towards edge computing. The method includes: collecting video data to be identified, and transmitting the video data to be identified to a first edge node connected to it, and according to the Calculate the joint point coordinate position of the video data, determine the skeleton data of the video data to be recognized, and process the skeleton data of the video data to be recognized by multiple second edge nodes based on different constraint strengths, obtain the human behavior recognition result of the video data to be recognized and uploading to the cloud server, and the cloud server receives the uploaded recognition result of the video data to be recognized, and fuses the human behavior recognition result to obtain the final human behavior recognition result. By increasing the number of edge nodes, more scale features are obtained, thereby improving the recognition accuracy. In addition, kinematics feature extraction and human behavior recognition are performed on edge nodes, which relieves network congestion and reduces the computing pressure on cloud servers.
Description
technical field [0001] The present application relates to the technical field of human behavior recognition, in particular to a method and system for extracting human behavior features oriented to edge computing. Background technique [0002] Human behavior recognition has important significance and broad application prospects in the field of security monitoring. Human behavior recognition in surveillance video is of great significance to social security and stability. The main task of human behavior recognition is to infer human behavior based on the data collected by visual sensors. The current human behavior recognition method can be divided into two steps: feature extraction and classification model construction: first, human behavior features are extracted from video data collected by visual sensors, and then a classification model is constructed to classify the features. [0003] In related technologies, the video data collected by the visual sensor needs to be uploa...
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): G06V40/20G06V20/40G06V10/80G06F9/50
CPCG06F9/5072G06F18/25
Inventor 王雪游伟
Owner TSINGHUA UNIV