华中科技大学学报（自然科学版）2011，Vol.39Issue(5)：109-112,4.
改进的基于区域合并的纹理图像分割方法
Improved textured image segmentation method by region merging
摘要
Abstract
According to the characteristics of natural textured images, a novel image segmentation method based on region merging was presented. The color feature of L* a* b* color space which accords with the color resolution of human visual perception was chosen. Then Gabor energy was extracted as a texture feature, and the similarity of two regions was measured by using the probabilityweighted average of L* a* b* color similarity and texture similarity. Finally, the maximal similarity based the region merging algorithm was used to partition the image with the guide of user inputs. The experimental results demonstrate that the method can obtain better the segmentation accuracy than those only utilizing RGB color feature. Under the same conditions of initial over-segmentation and user interaction, the method outperforms the currently popular image cutout tool-Lazy Snapping.关键词
纹理图像分割/颜色空间/Gabor能量/最大相似度/区域合并/人工交互Key words
texture image segmentation/ color space/ Gabor energy/ maximal similarity/ region merging/ user interaction分类
信息技术与安全科学引用本文复制引用
郑庆庆,桑农高,常鑫,刘乐元..改进的基于区域合并的纹理图像分割方法[J].华中科技大学学报（自然科学版）,2011,39(5):109-112,4.基金项目
国家高技术研究发展计划资助项目(2009AA12Z109) （2009AA12Z109）
教育部新世纪优秀人才支持计划资助项目(NNCET-05-0641). （NNCET-05-0641）
