基于BBO算法的二维交叉熵多阈值图像分割

Two-dimensional cross entropy multi-threshold image segmentation based on BBO algorithm

针对复杂图像的分割问题,提出一种基于生物地理学优化(biogeography-based optimization,BBO)算法的二维交叉熵多阈值图像分割方法。根据二维直方图斜分法得出交叉熵阈值选取公式,并将此推广到多阈值分割,以求得多个极值提高分割效果。由于二维交叉熵法在多阈值分割时计时长、复杂性高等问题,引入BBO算法的思想,实现对多个阈值快速精确的寻优。最后,对标准图像进行分割以验证该算法,结果表明此算法比二维交叉熵穷举法计算效率高。

To solve the problem of segmentation of complex images,this paper proposed a two-dimensional cross entropy multi-threshold image segmentation method based on biogeography-based optimization(BBO).First of all,it used cross entropy threshold selection formula based on two-dimensional histogram oblique method,which replaced the traditional thresholding formula to obtain multiple extreme values,so as to improve segmentation effect.Because of the high complexity in the process of multi-threshold segmentation,this paper introduced the BBO algorithm to realize the fast and accurate optimization of multiple thresholds.Finally,the improved algorithm in this paper segmented the standard image to verify the algorithm.The results show that the algorithm is more efficient than the two-dimensional cross entropy method of exhaustion.