一种改进的PCNN图像分割算法

An image segmentation algorithm based on an improved PCNN

PCNN用于图像分割时,为获得满意分割效果,其参数往往通过反复试凑确定,这在一定程度上限制了PCNN的使用。为此在改进的PCNN基础上,提出结合图像灰度直方图,以最大交叉熵函数作自适应遗传算法的适应度函数,采用自适应遗传算法搜索最优门限阈值的图像分割算法。该方法可有效地完成图像分割,分割结果优于原PCNN和传统Ostu算法。

PCNN is very suitable for image segmentation. However, it is necessary to determine the adaptive parameters of the network to achieve satisfactory segmentation results for different images. Up to now, the parameters of PCNN are always adjusted manually, which impedes its application in image segmentation. Aiming at the difficulties and shortcomings with using PCNN in image segmentation, combining the gray histogram of images, using the maximal cross-entropy function as the fitness function of Adaptive Genetic Algorithm, adopting Adaptive Genetic Algorithm to search the optimal threshold, an image segmentation algorithm is put forward based on the improved PCNN. This method can complete efficiently image segmentation, and the segmentation results are superior to the original PCNN and the traditional Ostu Algorithm.