基于二维直方图重建的Otsu图像分割算法  

Otsu Image Segmentation Algorithm Based on Rebuilding of Two-dimensional Histogram

作　　者：龚劬[1] 付云凤[1] 叶剑英[1] 姚玉敏[1] 

机构地区：[1]重庆大学数学与统计学院,重庆401331

出　　处：《计算机科学》2013年第8期313-315,共3页Computer Science

基　　金：中央高校基本科研业务费基金项目(CDJXS11100032)资助

摘　　要：针对二维Otsu算法因区域误分而产生的抗噪性差和计算量较大这一问题,提出了一种基于二维直方图重建的Otsu图像分割算法。该算法首先分析了原始算法中二维直方图所存在的误分和不足;然后重建二维直方图,以此来减弱噪声的干扰;最后将二维直方图区域划分由原来的四分法改为二分法,从而提高了计算速度。实验结果表明,本算法具有更强的抗噪性,分割效果也更为理想。Concerning the poor anti-noise performance in two-dimensional Otsu＇s method due to the obviously wrong region division and its huge calculation, this paper proposed an Otsu image segmentation algorithm based on rebuilding of the two-dimensional histogram. This method firstly analyzes the obviously wrong region division of two-dimensional histogram and its shortage, then rebuilds the two-dimensional histogram for reducing the noise interference, finally transfers the region division in two-dimensional histogram from four partitions into two partitions, thus reducing the search space of threshold,which saves a lot of processing time. The experimental results show that the proposed algo- rithm has better anti-noise performance and better segmentation result.

关 键 词：二维OTSU 图像分割 阈值 二维直方图 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]