Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A sleep behavior detection method based on deep learning

A detection method and deep learning technology, applied in the field of deep learning, can solve the problems of inability to adopt similarity detection, obvious difference in target size, affecting program judgment, etc., so as to reduce memory and time consumption, improve recognition ability, and improve accuracy. Effect

Active Publication Date: 2019-03-15
SHANGHAI JIAO TONG UNIV
View PDF5 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, there are several difficulties in the detection of students' sleeping behavior: 1. Real-time performance. Since the technology is based on video streams, there are high requirements for processing speed; 2. Scale variability. The size of the targets in the front row of the classroom is obviously different; 3. The characteristics are variable, and the sleeping postures are various. It is not possible to use a fixed template form for similarity detection.
4. Confusion, there are many common postures, such as bowing your head and writing, which may be very similar to sleeping postures, and there are many image noises that may affect the judgment of the program

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
  • A sleep behavior detection method based on deep learning
  • A sleep behavior detection method based on deep learning
  • A sleep behavior detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is carried out on the premise of the technical solution of the present invention, and detailed implementation and specific operation process are given, but the protection scope of the present invention is not limited to the following embodiments.

[0049] The present invention implements a sleep behavior detection method based on deep learning, cuts the video stream into frames, outputs discrete pictures, and then inputs the pictures of each frame into the convolutional neural network for feature extraction, and extracts the extracted Classify the features of the sleeping posture to detect the sleeping posture; use the efficient multi-scale detection technology to process the sleeping posture; finally use the tracking algorithm to track the sleeping students and judge whether they are sleeping or not.

[0050] The detailed techn...

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 a sleep behavior detection method based on deep learning, and the method comprises the following steps: 1) obtaining a to-be-detected video stream, and carrying out the framecutting processing to obtain a discrete image; 2) inputting the discrete pictures into a trained convolutional neural network model in sequence, and detecting to obtain a sleep posture preliminary detection result and corresponding confidence; 3) screening all sleep posture preliminary detection results based on a multi-scale detection method to obtain a sleep posture final detection result; And 4) according to the final detection result of the sleeping posture, judging whether a sleeping behavior exists or not by adopting a target tracking algorithm based on position information. Compared with the prior art, the method has the advantages that the feature fusion is adopted in the target detection model to improve the accuracy rate, and meanwhile, a sleeping behavior decision algorithm withhigh accuracy rate is adopted to avoid misjudgment of the sleeping behavior.

Description

technical field [0001] The invention relates to the technical field of deep learning, in particular to a method for detecting sleeping behavior based on deep learning. Background technique [0002] In modern video surveillance systems, a large number of useful human behavior analysis techniques have been developed. For example, in the driver monitoring system, the driver's fatigue is judged by analyzing the driver's facial behavior; in the intersection monitoring, the pedestrian behavior is analyzed and predicted to reduce the probability of car accidents. Student behavior analysis based on video streams in the classroom is of great help to teaching work. For example, it can count students' yawning behavior and sleeping behavior to evaluate the teaching effect, and analyze the behavior of different students to provide effective guidance to students. . [0003] However, there are several difficulties in the detection of students' sleeping behavior: 1. Real-time performance....

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): G06K9/00G06N3/04
CPCG06V40/20G06V20/40G06N3/045
Inventor 李文申瑞民姜飞米里亚姆·赖纳
Owner SHANGHAI JIAO TONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com