請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/22163
標題: | 體感辨識技術於年長者行動能力判斷之可行性研究 The Application of Body-movement Identification Technology in Judging the Agility of the Elderly |
作者: | Chi-Yao Liao 廖啟堯 |
指導教授: | 吳政鴻(Cheng-Hung Wu) |
關鍵字: | 肌少症,體感辨識裝置,長者,行走速度,抬腿角度,影像偵測, Sarcopenia,Body-movement Identification Device,Elderly,Walking speed,Leg lift angle,Image detection, |
出版年 : | 2018 |
學位: | 碩士 |
摘要: | 本研究為一種肌少症簡易檢測方法及系統,其係以體感辨識裝置的感應器擷取受測者行走的行走影像。體感辨識感應裝置運算晶片將行走影像轉換為3D跨步深度圖像。動作姿態辨識單元計跨步深度圖像跨步移動至最大距離的所需時間,以求出受測者的行走速度值。體態辨識裝置感應器擷取受測者原地踏步的側面影像。影像偵測感應裝置運算晶片將偵測到受測者抬腿的圖像。動作姿態辨識單元建立儲存有複數樣本圖像的影像資料庫,各樣本圖像具有不同抬腿角度所致之物體變形幅度的影像特徵,每一樣本圖像定義有一種抬腿座標值,令動作姿態辨識單元於影像資料庫中比對出符合抬腿圖像的樣本影像,再讀取出相應的夾角。當受測者之行走速度低於預設行走速度值及抬腿角度低於預設角度值時,則判定受測者為疑似肌少症患者,俾能實現自動化檢測肌少症之目的。 This study is a simple detection method and system for Sarcopenia, which uses the sensor of body-movement recognition device to capture the walking image of the tested person. The chipset of the body-movement sensing device converts the walking image into a 3D stepped depth image. The body-movement recognition unit calculates the time required for the maximum distance between the stride steps image to obtain the walking speed value of the tested person. The body-movement recognition device sensor captures the side-view image of the stepping at the same place of the tested person. The body-movement chipset will detect the heights of the stepping legged image of the tested person. The body-movement recognition unit creates an image database of massive multiple sample images. Each sample image has different image characteristics of its deformation amplitude of the tested person due to the leg lift angle. Each sample image defines a leg lift coordinate value. The body-movement recognition unit compares the sample images corresponding to the leg-lifting images in the image database and then reads the corresponding angles. When the tested person’s walking speed is lower than the preset walking speed value and the leg lift angle is lower than the preset angle value, the subject is judged to be a suspected Sarcopenia patient, so that the purpose of automated detection of Sarcopenia can be achieved. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/22163 |
DOI: | 10.6342/NTU201801915 |
全文授權: | 未授權 |
顯示於系所單位: | 工業工程學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-107-1.pdf 目前未授權公開取用 | 8.84 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。