Study on Hand Gesture Recognition using Spatio-temporal Slice
碩士 === 義守大學 === 資訊工程學系碩士班 === 99 === Traditionally, the video processing handles the whole screen. In most situations, video capture device has fix position, angle, and direction. Video image has time correlation in time domain. Spatio-temporal slice method can observe the variations in the region o...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2011
|
Online Access: | http://ndltd.ncl.edu.tw/handle/68719687972527288110 |
id |
ndltd-TW-099ISU05392022 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-099ISU053920222015-10-23T06:50:32Z http://ndltd.ncl.edu.tw/handle/68719687972527288110 Study on Hand Gesture Recognition using Spatio-temporal Slice 基於時空切片之手勢辨識研究 Ao-Yu Li 李翺宇 碩士 義守大學 資訊工程學系碩士班 99 Traditionally, the video processing handles the whole screen. In most situations, video capture device has fix position, angle, and direction. Video image has time correlation in time domain. Spatio-temporal slice method can observe the variations in the region of images at the time domain to determine the hand information, and reduce the processing time. In this study, we combine background subtraction and Sobel edge detection to get the human body information, and use the projections on X-axis and Y-axis and analyze the histogram statistics to obtain the body region. According to the body structure, shoulder is the center point of hand movement, so we can set the analysis lines. We then use temporal differencing to highlight the information of hand movement by analyzing the variation information on the analysis lines. Through this hand gesture recognition, the mouse or joystick can be replaced. Jyh-Horng Jeng Yih-Lon Lin 鄭志宏 林義隆 2011 學位論文 ; thesis 43 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 義守大學 === 資訊工程學系碩士班 === 99 === Traditionally, the video processing handles the whole screen. In most situations, video capture device has fix position, angle, and direction. Video image has time correlation in time domain. Spatio-temporal slice method can observe the variations in the region of images at the time domain to determine the hand information, and reduce the processing time. In this study, we combine background subtraction and Sobel edge detection to get the human body information, and use the projections on X-axis and Y-axis and analyze the histogram statistics to obtain the body region. According to the body structure, shoulder is the center point of hand movement, so we can set the analysis lines. We then use temporal differencing to highlight the information of hand movement by analyzing the variation information on the analysis lines. Through this hand gesture recognition, the mouse or joystick can be replaced.
|
author2 |
Jyh-Horng Jeng |
author_facet |
Jyh-Horng Jeng Ao-Yu Li 李翺宇 |
author |
Ao-Yu Li 李翺宇 |
spellingShingle |
Ao-Yu Li 李翺宇 Study on Hand Gesture Recognition using Spatio-temporal Slice |
author_sort |
Ao-Yu Li |
title |
Study on Hand Gesture Recognition using Spatio-temporal Slice |
title_short |
Study on Hand Gesture Recognition using Spatio-temporal Slice |
title_full |
Study on Hand Gesture Recognition using Spatio-temporal Slice |
title_fullStr |
Study on Hand Gesture Recognition using Spatio-temporal Slice |
title_full_unstemmed |
Study on Hand Gesture Recognition using Spatio-temporal Slice |
title_sort |
study on hand gesture recognition using spatio-temporal slice |
publishDate |
2011 |
url |
http://ndltd.ncl.edu.tw/handle/68719687972527288110 |
work_keys_str_mv |
AT aoyuli studyonhandgesturerecognitionusingspatiotemporalslice AT lǐáoyǔ studyonhandgesturerecognitionusingspatiotemporalslice AT aoyuli jīyúshíkōngqièpiànzhīshǒushìbiànshíyánjiū AT lǐáoyǔ jīyúshíkōngqièpiànzhīshǒushìbiànshíyánjiū |
_version_ |
1718110578808979456 |