基于自适应滤波的噪声图像边缘检测方法

研究图像边缘优化检测问题,针对传统边缘检测算法对噪声处理能力欠佳的缺陷,提出一种自适应中值滤波与形态学处理相结合的噪声图像边缘检测AMM算法。首先根据噪声像素点与相邻像素点的关联程度采用自适应中值滤波算法对图像进行去噪处理,保护图像的细节信息;然后运用改进的Canny算子对图像进行边缘检测,提取图像的边缘信息;最后对提取的边缘进行形态学处理,得到清晰的图像边缘。仿真结果对比表明,与目前常用的方法相比,AMM算法具有较强的抗噪鲁棒性,能较清晰地提取出图像的边缘,降低噪声对图像边缘的影响。

Edge Detection Algorithm Based on Self-adaptive Filter for Noise Image

To cope with the difficulty in digital image edge detection algorithms,a new and smart edge detection algorithm named AMM was proposed based on self-adaptive median filter,which improved Canny operator and mathematical morphology.Firstly,self-adaptive median filter was adopted to pre-processing image,better preserving image details and suppressing speckles.Then,a multi-scale edge detection algorithm was used to detect image edge by improved Canny operator.Finally,the image edges detected by Canny operator were processed by mathematical morphological filter.This AMM algorithm takes good advantages of self-adaptive filter and morphological filter.Simulation experiments show that the AMM algorithm has good effect in image edge detection and has better anti-noise performance than other methods.