An Image Contrast Enhancement Method bases on Fuzzy Set Theory

• Main Authors: Chung-ang Wang, 王中昂
• Other Authors: Chi-Farn Chen
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/15871397844238931757

碩士 === 國立中央大學 === 土木工程研究所 === 98 === The contrast of satellite images are usually affected by lots of factors. In order to increase the contrast, image enhancement techniques are the easiest methods. However, the darker and brighter area of original image could be compressed and lose detail information. Then user cannot see the detail of these area. In this study, we provide a fuzzy-based image enhancement method to compensate the brightness lost of the darker and brighter area of the image. There are three stages of the algorithm: First, classify the image by Fuzzy c-Means clustering method. Then we can get the membership value of each class of each pixel. Second, create enhancement model which is based on membership value of each class. Third, set membership value as the right and calculate the gray value of enhanced image. After getting the enhanced image, we evaluate the contrast by Michelson index and the quantity of information by Shannon entropy. Then compare the result and data with the traditional enhancement method. The result indicate that the proposed method could compensate the brighter and darker area and also provide an enhanced image with the same contrast as the traditional enhancement method.