分层遗传算法实现图像边缘检测

Edge Detection Design with Hierarchic Genetic Algorithms

提出一种基于分层遗传算法和梯度算子图像边缘检测的方法得到最优边缘算子,利用分层遗传算法优化设计灰度梯度算子,通过对样本图像的训练,在最小均方误差准则下获得最优灰度梯度算子。最优梯度算子增加了对噪声的抗干扰能力,并使得到的边缘更准确。

The operator optimized by hierarchic genetic algorithms is presented.We design the operator using hierarchic genetic algorithms,then optimize the operator under the least square criterion by training the sample images.The optimized operator has enhanced the repellent ability for noise that can get the edge more exactly under various conditions.