Urine Image Recognition For Medical Diagnosis Application

• Main Authors: Yu Pin Kao, 高煜彬
• Other Authors: W. S. Feng
• Format: Others
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/80242484286625301561

碩士 === 長庚大學 === 電子工程學研究所 === 96 === Since the urine examination is one of most important medical examination system, it shows the early inspection, and the early treatment. It can prevent from kidney-trouble problem or other relevant the only effective disease. But, in order to realize the urine phantom identification, the presented paper uses the vector calculating method in which J. Zou et al. mentioned to carry on the edge detection. First, carries on vector to perform algorithm picture, urine of sediment, have color picture elements that tracing change comes out to look for a bit picture, and set up the critical value, as well as divide the picture into the two-value picture. It can find out the edge of the sediment in the picture with our algorithm. To search the aspect in the area of sediment, the amount of horizontal and vertical projection with the rectangular picture of the projection income can be calculated. Finally, the sediment area and their quantity in the image are obtained.