CAJ | 학술논문

针对传统图像增强过程中产生的伪吉布斯现象，清晰度差和对比度低的问题，提出一种基于非F采样Contourlet变换（NonsubsampledContourletTransform，NSCT）的反锐化掩模图像增强算法。该算法一方面利用NSCT变换的平移不变性，抑制传统增强算法中产生伪吉布斯现象，另一方面用反锐化掩膜算法来处理图像，提高增强后图像的清晰度和对比度。实验仿真结果表明，本文提出的方法与Conlourlet增强方法，反锐化掩模增强方法相比能够有效提高图像的清晰度，对比度和抑制伪吉布斯现象，图像视觉效果有明显改善。
침대전통도상증강과정중산생적위길포사현상，청석도차화대비도저적문제，제출일충기우비F채양Contourlet변환（NonsubsampledContourletTransform，NSCT）적반예화엄모도상증강산법。해산법일방면이용NSCT변환적평이불변성，억제전통증강산법중산생위길포사현상，령일방면용반예화엄막산법래처리도상，제고증강후도상적청석도화대비도。실험방진결과표명，본문제출적방법여Conlourlet증강방법，반예화엄모증강방법상비능구유효제고도상적청석도，대비도화억제위길포사현상，도상시각효과유명현개선。
In order to solve pseudo- Gibbs phenomenon, low contrast and blurred phenomenon in the process of image enhancement, a new algorithm is proposed which is in basis of NSCT domain. On the one hand, this proposed algorithm makes use of NSCT' s shift - invarianee to prohibit the pseudo - Gibbs phenomenon, on the other hand, deals with NSCT's coefficient by Unsharp Masking to improve contrast and get better visual effect. Compared with the Contourlet method and Unsharp Masking method, the proposed method is obviously superior both in contrast and in vision.