摘要
针对迭代盲反卷积算法恢复图像在内部灰度对比鲜明的地方、图像边界都出现的"振铃"效应和迭代盲反卷积算法通常去模糊效果好而去噪效果不理想等情况,本研究提出两点改进:首先,迭代盲解卷积算法恢复图像之前对模糊带噪的退化图像进行改进的中值滤波去噪;其次,对恢复出来的图像,通过找到内部灰度变化大的部分和边界并把其像素设置为0,进行去除"振铃"效应.改进后的算法简单,不仅复原了图像的细节,而且很好地去除了噪声.实验结果表明,改进算法取得了比较好的复原效果.
Two methods have been introduced to resolve the so- called ring phenomena near edges effectively and stark contrast gray on the inside and iterative blind deconvolution （IBD）can get a good deblurring result but obtain bad denoising effect. Firstly,To de- noise blurred and noise degradation image using an improved median filtering algorithm before IBD algorithm were processed for image processing; secondly, banishes the so called ringing phenomena by finding near edges and stark contrast gray and setting the pixels zero. This method is simple, it not only keeps more contours and textures in the reconstructed images but denoises well. The experimental results showed that the method has been quite good results.
出处
《周口师范学院学报》
CAS
2011年第2期52-54,共3页
Journal of Zhoukou Normal University
