A Study on Application of Images Recognition to Protect Human Eyes of Computer User

• Main Authors: Wei-Wen Chen, 陳威文
• Other Authors: 黃榮堂
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/mcrcuk

碩士 === 國立臺北科技大學 === 機電整合研究所 === 93 === This study developed a system for protecting computer user’s eyes by applying machine vision techniques such as face detection, face recognition, visual distance detection, and ellipse matching, etc. Regarding the face detection, a color self-calibration algorithm is employed to eliminate the influence of color image caused by the light source of environment. Then, skin extraction and creating a skin window are followed to find the skin region of the image. For searching the skin region real-time, we choose algorithm based on vertical projection instead of labeling. Each skin region has to be checked via ellipse matching to identify human face so as to complete the process of face detection. In addition we also detect the busy signal of the keyboard and mouse to guarantee the detection of the real computer using time of the user. Regarding the visual distance, the distance is detected by the width of user’s face from ellipse matching. Blinking detection is using eyes template matching and XNOR to search the existence of available open eyes. However, it can''t achieve the goal of distinguishing personages yet in this way. Why distinguish personages? Because when the user leaves for a while, then comes back and continues using computer we have to judge whether he is the same user or not. Accordingly, service time of the computer should be continued totally or should be reset, will depend on the face recognition. The face recognition is using eigenface method in this research. For the purpose of protecting the computer user’s eyes, eigenface method has precise recognition rate enough to distinguishing personages.