融合特征自适应抑制式模糊聚类彩色图像分割

Adaptively suppressed fuzzy clustering color image segmentation with fused features

为了提高彩色图像分割的精度和效率,提出了一种融合特征自适应抑制式模糊聚类图像分割算法。在Lab空间提取图像色彩信息,采用Haar小波变换与半方差函数提取图像纹理特征,得到7维融合特征以概括图像信息。利用带宽自适应的均值漂移算法生成聚类数目和初始聚类中心。根据迭代过程中隶属度的动态变化自适应生成抑制因子,以改善算法的运行效率。仿真结果表明,与相关经典算法相比,改进算法的分割精度较好,运行效率较高。

In order to improve the accuracy and efficiency of color image segmentation,an adaptively suppressed fuzzy clustering image segmentation algorithm with fused features is proposed.The image color information is extracted in Lab space,the image texture features are extracted by Haar wavelet transform and semi-variance function,and the obtained 7-dimensional fusion feature is used to summarize image information.A bandwidth adaptive mean-shift algorithm is adopted to generate the number of categories and initial clustering centers.According to the dynamic change of membership in the iterative process,an adaptive generation formula of inhibitory factor is constructed to improve the efficiency of the algorithm.Simulation results show that compared with the related classical algorithms,the improved algorithm has better segmentation accuracy and higher operation efficiency.