基于小波和模糊聚类的示温漆彩色图像分割算法

Algorithm of Thermopaint Color Image Segmentation Based on Wavelet Transform and Fuzzy Clustering

针对小波分析和聚类算法用于彩色图像分割的不足,提出了一种结合小波和模糊聚类示温漆彩色图像分割算法.首先通过小波变换模值得到图像各颜色分量的模值,并与颜色特征一起作为颜色特征量;然后根据聚类算法计算欧氏距离;最后根据欧氏距离和颜色特征量,按照模糊C均值聚类算法完成图像分割.实验结果证明,将该方法应用于示温漆彩色图像分割中,取得了较好的效果.

An algorithm of thermopaint color image segmentation combined with wavelet analysis and fuzzy clustering is proposed in view of deficiencies between wavelet analysis and clustering algorithm. Firstly,each color component modulus value is acquired through wavelet transform modulus values,and is used for color eigenvalue together with color characteristics. Secondly,Euclidean distance calculated out according to fuzzy clustering algorithm. Finally,image segmentation is accomplished through fuzzy C-means algorithm according to Euclidean distance and eigenvectors. Experimented results show that this method has good effect in thermopaint color image segmentation.