Defect Detection on Aged Films

• Main Authors: Hsing-Ying Chung, 鍾興穎
• Other Authors: 施國琛
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/13563115654492008037

碩士 === 淡江大學 === 資訊工程學系碩士班 === 96 === Aged motion pictures may contain different types of defects, such as spikes and scratch lines. Each type of defect has its different properties. It is relatively hard to precisely detect these defects on aged films by using only a single solution. This paper proposes two techniques to detect spikes and scratch lines on aged films. In detecting spikes, we use the motion estimation and morphological image processing. In detecting scratch lines, we propose to compute the mean of intensity on each column, and then use the standard deviation and compute the number of the weight to obtain the result of detecting.