嬰幼兒姿態偵測演算法之開發及其照護系統之應用;The Development of Infant and Children Posture Detection Algorithms and Its Applications in Healthcare Systems

NCU Institutional Repository > 資訊電機學院 > 資訊工程學系 > 研究計畫 > Item 987654321/84714

請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/84714

題名:	嬰幼兒姿態偵測演算法之開發及其照護系統之應用;The Development of Infant and Children Posture Detection Algorithms and Its Applications in Healthcare Systems
作者:	蘇木春
貢獻者:	資訊工程學系
關鍵詞:	姿態辨識系統;學習機器;類神經網路;嬰兒風險檢測;posture detection system;learning machine;neural networks;infant risk detection system
日期:	2020-12-08
上傳時間:	2020-12-09 10:46:25 (UTC+8)
出版者:	科技部
摘要:	此計畫是一個三年期的計畫，整體的總目標是要開發嬰幼兒的姿態偵測系統及其於嬰幼兒不同時期的照護系統的開發。臨床上，要精準判斷一個嬰兒是否在運動神經發展上有所受損，不是依賴一個有經驗的醫生在一段錄影好的影片上利用視覺來評估，就是得要仰賴昂貴的儀器。可想而知，這是個耗時且昂貴的診療程序。所以，一個能夠自動化評估的輔助系統就變得十分迫切了。最近幾年來，OpenPose 是一個有效且被普遍採用的姿態估測的演算法，但它的缺點是它是針對成人姿態所訓練而成的，導致若是應用於嬰兒姿態偵測時，整體效果會下降。所以，在第一年的計劃中，我們會開發一個嬰兒姿態偵測演算法，然後，據此演算法開發出「智慧型早產兒風險檢測系統」。　　第二年的計劃中，我們會將此嬰兒姿態偵測演算法用來開發： (1) 「智慧型嬰兒危險偵測系統」和 (2) 「智慧型嬰兒動作與音樂互動系統」。若躺在嬰兒床的嬰兒有一些危險動作產生 (如：久睡不動、異物塞鼻或吐奶等)，則「智慧型嬰兒危險偵測系統」會發出警示信號給照護者；而「智慧型嬰兒動作與音樂互動系統」則是要要讓嬰兒可以透過肢體的動作來觸發嬰兒床邊的音樂鈴，讓嬰兒探索身體與音樂的互動性。　　第三年的計畫目標對象則是針對幼兒來開發：(1)「智慧型學習姿態辨識系統」和 (2)「智慧型幼兒危險警示系統」。「智慧型學習姿態辨識系統」可以掌握孩童念書或上課時的專注的程度，並能適時提醒孩童眼睛是否距離書本太近或是有不良之閱讀坐姿（如：駝背、手撐頭等）；「智慧型幼兒危險警示系統」則在一些危險動作發生時 (如：打架、爬高後跳下、將異物體放入口中等)，可以適時提醒父母親或幼稚園老師以掌握孩童活動的狀態。其目標是要減輕父母親或照護者的負擔和掌握孩童學習的狀態。 ;This is a three-year project. The main goal of this project is to develop algorithms for detecting postures of infants and children and then apply the posture detection algorithms to implement health caring systems for different periods of infants and children. Clinical assessments for infant’s risk of developing neuromotor impairment either are assessed through visual examination by specialized clinicians via recorded videos or involve expensive equipment, which are usually time-consuming, expensive, and only available in highly-resourced environments. This makes assessment inaccessible for families of limited means and in low resource countries. Therefore, it is desirable to automate the process of evaluating the quality of infant movements; otherwise, the early identification is not possible. In recent years, the OpenPose is a very popular human pose estimation algorithm; however, it focuses on adults, leading a degradation of accuracy if applied to infants. Therefore, In the first year, we will fist develop a pose estimation algorithm especially for infants and then apply it to develop an intelligent premature infant risk detection system for automatically assessing infant neuromotor risks.　　In the next year's plan, we will use this infant pose detection algorithm to develop (1) an intelligent baby danger detection system and (2) an intelligent baby music interactive system. If the baby lying in the crib has some dangerous movements (such as long sleep, stuffy nose and vomiting, etc.), the " intelligent baby danger detection system " will send a warning signal to the caregiver. The " intelligent baby music interactive system " is to allow the baby to trigger the music bell beside the crib through the movement of the limbs, so that the baby can explore the interaction between its body movements and music.　　In the third year's plan, we will develop (1) an intelligent reading posture recognition system and (2) an intelligent child's danger warning system for children. The intelligent reading posture recognition system can fully monitor the degree of concentration of children in their studies and promptly remind them whether the eyes are too close to the books or they have poor reading postures (e.g., humpback, one hand supporting his/her head). The intelligent child's danger warning system can prompt parents or kindergarten teachers to notice the status of children's activities when some dangerous actions (such as fighting, jumping after climbing high, putting objects into the mouth, etc.) occur. The goal of these two systems is to reduce the burden on parents and master the status of children learning.
關聯:	財團法人國家實驗研究院科技政策研究與資訊中心
顯示於類別:	[資訊工程學系] 研究計畫

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
index.html		0Kb	HTML	200	檢視/開啟

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....