改进模糊逻辑的边缘检测算法

Improved Fuzzy Logic Edge Detection Algorithm

针对Pal&King模糊边缘检测算法中参数难以选定,低灰度图像信息缺失,公式复杂,反复迭代耗时且效果不确定等缺点,提出基于模糊理论的边缘检测改进算法.首先通过衡量像素邻域内的相关性进行模糊化;然后用Sugeno模糊模型进行模糊推理,增强边缘点,弱化区域点;最后用简单的解模糊过程.改进算法使模糊化和解模糊过程中的运算更简单,省去Pal&King算法中的多次迭代和各种参数和阈值的设定.结果表明,该算法可以提高Pal&King算法的效率,使检测效果进一步增强,具有更好的通用性.

Regarding the shortcomings of hard-to-select-parameter,shortage of low grayscale image information,complex formula,time-consuming and uncertain consequence of repeated iterations in Pal King fuzzy edge detection algorithm,the article raises improved edge detection algorithm based on fuzzy theory.First of all,it fuzzed by measuring the correlation in pixel neighbourhood;then fuzzed reasoning with Sugeno Fuzzy Model to enhance the boundary point and weaken the zone point;finally it adopted simple de-fuzzification process.The improved algorithm makes the calculation process of fuzzification and de-fuzzification more simple,leaving out the multi-iteration and the setup of various parameters and threshold values in Pal King algorithm.The result shows that,the improved algorithm can increase the efficiency of Pal King algorithm,and enhance the detection effect,performing better generality.