A Real-Time Automatic Color Image Enhancement

• Main Authors: Feng-Shen Weng, 翁豐盛
• Other Authors: Nai-Jian Wang
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/vmvhg4

碩士 === 國立臺灣科技大學 === 電機工程系 === 99 === In the digital photographing, the quality of the images may be affected due to the lack or the changes of the ambient light. In the past, many image enhancement schemes for the image enhancement are proposed. However, those schemes focus on either the brightness or the contrast of the image. In the thesis, proposed method to be proposed a scheme for automatic image enhancements, including the contrast enhancement, the brightness enhancement, and the image classification. First, the image classification categorize input images according to the standard deviation and the average luminance, and then take the corresponding schemes to enhance the contrast or the brightness of the images. In the luminance enhancement, the propose method modified the scheme of the color image enhancement based on the space variant luminance map, which is proposed by Lee et al. In the contrast enhancement, our propose method modified the scheme based on the weighted histogram equalization, which is proposed by Yun et al. Proposed method overcome the challenge that the input image is not easy to enhance when the input image contrast is extreme poor condition. The proposed scheme of automatic image enhancements is validated by a Borland C++ program. It should be noted that the quality of images is better than that of the schemes proposed by Yun et al. and Lee et al.