摘要： 在对舌图像的去噪过程中，平滑噪声的同时容易丢失边缘和纹理等细节信息。为此，研究基于偏微分方程的舌图像去噪方法，分别采用中值滤波、高斯滤波、P-M方程、正则化P-M方程以及耦合冲击-复扩散滤波模型，对加噪舌图像进行滤波。比较结果表明，正则化P-M方程更适合舌图像的去噪处理，该方法处理速度快、去噪效果好，且能有效保护图像边缘。
关键词:
舌诊,
偏微分方程,
图像去噪,
P-M方程,
冲击滤波,
复扩散
Abstract: When denoising the tongue image, the details such as the edges and the veins are easy to be lost. Aiming at this problem, this paper proposes a new method to denoise the tongue image based on Partial Differential Equation(PDE). It applies median filtering, Gaussian filtering, P-M equation, regularization P-M equation and Coupling shock filtering combined complex diffusion filtering model to denoise the noising tongue images. Comparison result shows that the regularized P-M equation is the best method to denoise the tougue images. It has fast speed and good denoising result, and the best preserving for edges.
Key words:
tongue diagnosis,
Partial Differential Equation(PDE),
image denoising,
P-M equation,
shock filtering,
complex diffusion
中图分类号:
唐苏湘, 汪仁煌, 韦玉科, 明俊峰, 何最红. 基于偏微分方程的舌图像去噪研究[J]. 计算机工程, 2012, 38(14): 190-192.
TANG Su-Xiang, HONG Ren-Huang, HUI Yu-Ke, MENG Dun-Feng, HE Cuo-Gong. Research of Tongue Image Denoising Based on Partial Differential Equation[J]. Computer Engineering, 2012, 38(14): 190-192.