用于灰度图像分割的小波区域扩张算法  

Gray Image Segmentation Based on Wavelet Region Growing Algorithm

作　　者：刘旭红[1] 张国英[1] 刘玉树[1] 柳晓雷[1] 

机构地区：[1]北京理工大学计算机科学与工程系,北京100081

出　　处：《计算机工程》2004年第22期132-133,194,共3页Computer Engineering

摘　　要：文献[1,6]提出了小波变换聚类(WaveCluster)算法。该文将这种思想引入区域扩张的图像分割算法中,提出了小波区域扩张算法,首先对图像进行小波变换,在变换空间中进行图像分割,再将分割结果映射到原始空间,有效地提高了分割效率。The literatures[1,6] propose an efficient algorithm WaveCluster. This article introduces this idea to the image segmentation based on region growing and produces a new algorithm——wavelet region growing. Firstly the wavelet transform is appliecl to the image, then the image is segmented in the transformed space, finally the result is mapped to the original image space. The efficiency of segmentation is improved efficiently.

关 键 词：图像分割算法 图像编码 灰度图像 小波区域扩张算法 尺度函数 

分 类 号：TN919.81[电子电信—通信与信息系统]