自适应烟花算法下多维模糊C均值彩色图像分割

针对图像分割在自然场景中,分割精度不高和细节保持不够敏感,提出一种自适应烟花算法下的多维模糊C均值彩色图像分割算法。结合动态时间弯曲思想,以邻域像素相似特点构造弯曲曲线,得到多维相似距离和新的目标函数。在自适应烟花寻优算法下,找到最优聚类中心,最终达到对图像分割效果。实验表明,该算法与同类算法相比,对彩色图像有良好的分割效果,对图像的细节保持也不错。

MULTIDIMENSIONAL FUZZY C-MEANS COLOR IMAGE SEGMENTATION BASED ON ADAPTIVE FIREWORKS ALGORITHM

Aiming at the shortcomings of euclidean distance in traditional fuzzy C-means clustering algorithm,as well as the determination of clustering center and iteration calculation,a color image segmentation method based on multi-dimensional fuzzy C-means in adaptive fireworks algorithm is proposed.Combined with the idea of dynamic time warping,the warping curve was constructed based on the similarity characteristics of domain pixels,and the multi-dimensional similarity distance and new object function were obtained.Under the adaptive fireworks optimization algorithm,the optimal clustering center was found,and finally the image segmentation effect was achieved.The experiments show that the proposed algorithm has a good segmentation effect on color images and good details preservation compared with other similar algorithms.