- 學位論文
手掌穴道辨識之擴增實境系統開發
On the Development of an Augmented Reality System for Palm Acupoint
摘要
穴道的運用可視為中國傳統醫學的一種治療方式，可廣泛應用在日常生活的保健上。然大部分人們對於穴道的位置以及相關知識並不太熟悉，而市面上的書籍或是網路上的資訊，大多都是以圖文來呈現與說明，讓人們不是很輕易理解。因此本研究開發了一項系統進行手掌穴道的辨識，主要辨識手掌正反面共23個穴道，並透過擴增實境來顯示手掌穴道的位置以及讓使用者進行互動，以此可以更加方便和快速地找到穴道。 本系統屬於無標記式擴增實境，是一款開發於Android平台上執行的App，開發上可分為三個部份，(1)影像辨識，經由動態影像中偵測影像手掌目標；(2)手掌穴道搜尋，運用平面座標系之向量比例尺的概念，即時判斷影像手掌穴道的位置；(3)擴增實境的顯示與使用者互動，呈現穴道的位置以及相關圖文資訊。本研究以30位受測者進行實驗測試，使用SUS易用性量表進行評估的結果分數為81.25分，是屬於中上等級的使用性評價。另外根據本研究提出的同心圓距推測法，檢測本系統辨識的穴道準確度，其結果顯示，若穴道是分佈在手掌輪廓內，準確度範圍平均是落在70%-79%之間的級距；而如果是分佈在手掌輪廓邊緣上的穴道，平均準確度範圍是位於52%-70%的級距。由此兩類標準的數據推論，每位受測者的手掌穴道之準確度範圍幾乎都有達到中上的標準，代表本系統所顯示的穴道點是相當吻合於實際穴道的位置。
並列摘要
Acupressure is popular way to perform therapy in the traditional Chinese medicine (TCM). It has been widely used in various method of health care. Currently, to learn the position of the acupoints for applying acupressure, most people acquire the knowledge of the acupoints through books on the market or and information on the web. Although the knowledge is presented with text and graphic illustration, it is still uneasy to understand for many people. In this study, we have applied various cutting-edge technology to develop a system for learning the position of the acupoints. The system is an App on mobile devices with Android platform. Through the Augmented Reality, this system display the acupoints and allows users to interact with the system. It can identify twenty-three acupoints of the palm. This system adopts Markerless Augmented Reality and uses digital image processing technology to develop the identification of these acupoints of human palm. The AR presentation and the direct interaction allow users to quickly gain familiarity with information about acupoints. The technical merits of our App can be described in the following three parts: (1) to detect palm from motion picture by using digital image processing technology, (2) to search acupoints of palm by applying the concept of vector plane coordinate system of scale to judge the position of the acupoints of palm in real-time, (3) to interact with AR environment through developing software to facilitate the locating of acupoints and the touching of screen to display relevant information on the mobile device. To do the field test of the App, we employed the SUS (System Usability Scale) to do usability study. We convened thirty users to participate the user study. The result is quite good with a SUS survey score 81.25. The feedback from these users is explored and discussed for further improvements of the App.