一种改进的自适应图像去噪算法实现

An improved adaptive image de-noising algorithm and its implementation

数字图像在形成、传输和处理过程中,往往不可避免的存在噪声污染。因此去噪是数字图像处理的重点也是难点,需要提出一种更加行之有效的去噪算法。针对于传统软硬阈值的缺陷,本文提出一种将改进的的小波阈值去噪法与自适应中值滤波相结合的新的去噪方法。小波阈值去噪即确定一个合适的阈值,对分解的小波系数进行阈值化处理,滤除噪声的小波系数,而保留信号的小波系数。该算法相对于传统软硬阈值的峰值信噪比(PSNR)的提高百分比分别达到了16.32%和8.95%,去噪效果非常理想。该方法较之以往的去噪方法,峰值信噪比有了明显的提高,同时能够有效保留边缘及细节特征。

In the process of digital image formation,transmission and processing,noise pollution is inevitable. Therefore,de-noising is the focus on digital image processing,is also difficult,It needs to propose an improved effective de-noising algorithm. Aiming at the defects of the traditional soft and hard thresholds,a new de-noising method based on improved wavelet threshold de-noising method combined with adaptive median filter is proposed in this paper. Wavelet threshold de-noising is to determine a suitable threshold,the decomposition of the wavelet coefficients of the thresholding,filters out the noise of the wavelet coefficients to retains the signal. Compared with the traditional soft and hard thresholds,the improved peak signal to noise ratio（ PSNR） of the algorithm is 16. 32% and 8. 95%,respectively,and the de-noising effect is quite ideal. Compared with the conventional de-noising method,the peak signal to noise ratio is improved obviously,and the edge and detail features can be effectively preserved.