摘要
为改善消噪后图像的质量,提出一种在保留图像细节信息的同时,能够消除污染图像脉冲噪声的有效算法.该算法采用模糊噪声检测技术,结合差分绝对值顺序(ROAD)统计量和开关函数构造模糊噪声检测器,采用像素原始值、中值和像素噪声疑似度的线性组合消除噪声.该算法不需预先训练,通过脉冲噪声检测和消除两个步骤即可消除图像脉冲噪声.实验结果表明,该算法在峰值信噪比和视觉效果方面优于其它图像消噪算法.
In this paper, a novel efficient algorithm is proposed, to improve the quality of the image removable noise, which can remove impulse noise from images. It uses a fuzzy impulse detection technique to combine rank-ordered absolute differences (ROAD) statistics with a binary value function to create a fuzzy impulse noise detector. The restored pixel value is a linear combination of orig' inal pixel value , pixel median value and pixel noise doubt modality. It requires no previous training, which removes impulse noise from corrupted images via two steps of impulse noise detection and impulse noise cancellation. Experimental results show that the approach is better than other algorithms for image noise removal on peak signal-to-noise ratio (PSNR) and vision.
出处
《沈阳理工大学学报》
CAS
2009年第1期51-53,共3页
Journal of Shenyang Ligong University
关键词
图像恢复
ROAD统计量
模糊噪声检测
脉冲噪声消除
image restoration
ROAD ( rank-ordered absolute differences) statistic
fuzzy noise detection
impulse noise cancellation