【作者】 李兰兰； 吴乐南；

【Author】 Li Lanlan Wu Lenan Department of Radio Engineer, Southeast University, Nanjing 210096, China

【机构】 东南大学无线电工程系；

【摘要】 许多早期的视觉处理过程在去除图像噪声的同时还锐化图像边缘。由Perona和Malik提出的偏微分方程能自适应的平滑图像噪声,P-M方程在图像灰度变化不大区域的扩散速度较快,在边缘区域扩散速度慢。基于冲激滤波模型的图像增强算法能够锐化图像边缘。本文在各向异性扩散和冲激滤波的基础上提出了一种对带噪声文本图像的增强算法。通过实验证明了该方法的有效性。更多还原

【Abstract】 The goal of many early visual filtering processes is to remove noise while at the same time sharpen edge. The nonlinear anisotropic diffusion proposed by Perona and Malik can adaptively control the smoothing by encouraging intra-region smoothing while inhibiting inter-region interactions. Shock filters can also enhance edge of image. In this paper, an enhancement algorithm for noised document image based on anisotropic diffusion and shock filters is presented. Experiment result shows that this method is valid.更多还原