Computer Vision Techniques for Detection and Recognition of Drinking Activity

This thesis presents two novel computer vision techniques for detection and recognition of drinking activities at home which utilise only the depth information from RGBD cameras. According to my best understanding, there is very little work on using video cameras with depth sensor for the detection...

全面介绍

Saved in:
书目详细资料
主要作者: Tham, Jie Sheng
格式: Thesis
出版: 2016
主题:
标签: 添加标签
没有标签, 成为第一个标记此记录!
id my-mmu-ep.7149
record_format uketd_dc
spelling my-mmu-ep.71492018-05-21T15:40:28Z Computer Vision Techniques for Detection and Recognition of Drinking Activity 2016-08 Tham, Jie Sheng QA75.5-76.95 Electronic computers. Computer science This thesis presents two novel computer vision techniques for detection and recognition of drinking activities at home which utilise only the depth information from RGBD cameras. According to my best understanding, there is very little work on using video cameras with depth sensor for the detection and recognition of ambient assisted living dining activity. The main advantage of using depth information is that the accuracy will not be affected by the change of lighting condition and illumination, as compared with using the conventional RGB cameras. In particular, the first proposed technique extracts the features from the depth information of hand action characteristic during the drinking. As the drinking action features are gathered, dynamic time warping algorithm is used to recognise and detect the drinking activity. The experimental results show that the proposed method has a comparatively high recognition accuracy of 89% in comparison with the existing visual-based techniques. 2016-08 Thesis http://shdl.mmu.edu.my/7149/ http://library.mmu.edu.my/diglib/onlinedb/dig_lib.php masters Multimedia University Faculty of Engineering
institution Multimedia University
collection MMU Institutional Repository
topic QA75.5-76.95 Electronic computers
Computer science
spellingShingle QA75.5-76.95 Electronic computers
Computer science
Tham, Jie Sheng
Computer Vision Techniques for Detection and Recognition of Drinking Activity
description This thesis presents two novel computer vision techniques for detection and recognition of drinking activities at home which utilise only the depth information from RGBD cameras. According to my best understanding, there is very little work on using video cameras with depth sensor for the detection and recognition of ambient assisted living dining activity. The main advantage of using depth information is that the accuracy will not be affected by the change of lighting condition and illumination, as compared with using the conventional RGB cameras. In particular, the first proposed technique extracts the features from the depth information of hand action characteristic during the drinking. As the drinking action features are gathered, dynamic time warping algorithm is used to recognise and detect the drinking activity. The experimental results show that the proposed method has a comparatively high recognition accuracy of 89% in comparison with the existing visual-based techniques.
format Thesis
qualification_level Master's degree
author Tham, Jie Sheng
author_facet Tham, Jie Sheng
author_sort Tham, Jie Sheng
title Computer Vision Techniques for Detection and Recognition of Drinking Activity
title_short Computer Vision Techniques for Detection and Recognition of Drinking Activity
title_full Computer Vision Techniques for Detection and Recognition of Drinking Activity
title_fullStr Computer Vision Techniques for Detection and Recognition of Drinking Activity
title_full_unstemmed Computer Vision Techniques for Detection and Recognition of Drinking Activity
title_sort computer vision techniques for detection and recognition of drinking activity
granting_institution Multimedia University
granting_department Faculty of Engineering
publishDate 2016
_version_ 1747829653830107136