QQ客服 官方微博 反馈留言

	A Robust Optimum Thresholding Method Based on Local Intensity Mapping and Class Uncertainty Theory
	Yuntao Wang; Guoyuan Liang; Sheng Huang; Can Wang; Xinyu Wu; Yachun Feng
	2017
会议地点	中国澳门
英文摘要	The paper proposed a robust optimum thresholding method based on local intensity mapping(LIM), class uncertainty and region stability theories to segment fuzzy and noisy images. First of all, the intensities of an image would be mapped into another intensity space by LIM which could decrease the influence of noise and uneven intensity distribution. Then, the intensity-based class uncertainty is applied as the measurement of statistical information while the region stability is used for evaluating the spatial information embedded in image. After that, an energy function is constructed to search for the optimum threshold based on the class uncertainty and region stability. Experimental results demonstrate the proposed method can achieve better performance as well as robustness compared with some classical methods, such as Otsu, MHUE(minimization of homogeneity- and uncertainty-based energy) and Sauvola.
语种	英语
内容类型	会议论文
源URL	[http://ir.siat.ac.cn:8080/handle/172644/11882]
专题	深圳先进技术研究院_集成所
作者单位	2017
推荐引用方式 GB/T 7714	Yuntao Wang,Guoyuan Liang,Sheng Huang,et al. A Robust Optimum Thresholding Method Based on Local Intensity Mapping and Class Uncertainty Theory[C]. 见:. 中国澳门.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。