基于直方图特性的图像去噪方法

传统的图像去噪方法在去噪过程中往往导致处理过后的图像模糊不清,且残留的信号噪声较大.为有效去除噪声又不至图像模糊,本文针对多幅图像加噪声后的直方图进行了分析,提出了一种利用图像直方图特性去除噪声的算法,对加入噪声后的直方图进行扫描和灰度值变换以调整原噪声图像的灰度,使得噪声融入背景.最后利用中值滤波进行残留噪点的消除.实验结果表明,该算法不仅可以使得图像能有效去除噪声,而且图像的质量有较大幅度的提高.

Image denoising based on histogram

Traditional image denoising methods usually cause a blurred image after the image processing, and there is still a lot of residual noises. To get a denoising but not blurred image, this paper analysed many histograms of noise polluted images and presented a novel denoising algorithm based on the histogram characteristic. First scans the noise polluted histogram, then adjusts the gray value by gray value transformating to make the noise fusing into the image. At last, uses the median filter to complete the processing. Then a denoising and high quality picture is acquired. The experimental results show that the algorithm presented in this paper is feasible and good.