针对噪声多小波分解后的尺度性与图像本身的特性,通过遗传算法自适应寻求图像的最小均方误差,提出了一种基于遗传算法的多小波自适应阈值去噪算法。实验结果证明,该算法明显优于传统算法,不仅能有效地滤除图像中的噪声,而且能较好地保留图像的边缘信息,具有更为理想的去噪效果。
In view of the scaling of noise after muhiwavelet decomposed and characteristic of image, adaptive to look for the least RMSE of image according to genetic algorithm, we propose a multiwavelet multilevel threshold denoise method based on generation algorithm. The experimental result shows that this algorithm is superior to the traditional methods,it not only can remove the noise of image, but can remain the better edge of image, and has more perfect denoise effect.