摘要
每种图像分割方法都只利用了图像信息中的部分特征,必然带有局限性,因此只能针对各种实际应用领域的需求来适当选择所需的方法。比较了几种阈值分割和边缘检测算法,着重研究了最大类间方差算法,并对其进行改进。针对不同的图像进行了仿真,对实验结果进行了分析、研究、比较。结果表明,改进的Otsu算法能有效地提高图像分割的质量。
Some characteristics were used in the image segmentation algorithms, these algorithms have the inevitable limitations, so the appropriate methods were selected for a variety of practical applications in the field of demand. Some threshold segmentation, edge detection algorithms were discussed, the Otsu algorithm and the improved Otsu algorithm were studied. The results of the simulation by different images were analysed, studied, compared. The results show that the improved Otsu algorithm can effectively improve the quality of image segmentation.
出处
《计算机科学》
CSCD
北大核心
2009年第5期276-278,290,共4页
Computer Science
基金
重庆市科委自然科学基金计划资助项目(No.CSTC2007BB2451)资助
