基于模糊形态学的激光图像处理算法研究  

Research on algorithm of Laser image processing based on fuzzy morphology

作　　者：黄戎[1] 李智[2] 杨小义[3] 肖超[4] 

机构地区：[1]重庆工业职业技术学院自动化系,重庆401120 [2]重庆医科大学附属第二医院放射科,重庆400010 [3]重庆师范大学教育科学学院,重庆400030 [4]重庆大学自动化学院,重庆400044

出　　处：《激光杂志》2013年第3期26-28,共3页Laser Journal

摘　　要：针对原始激光图像辨识处理中存在边缘不清晰等问题,提出了一种基于模糊形态学的新图像分割算法。文中剖析了浸没模拟分水岭算法与IFT分水岭算法,对模糊形态学的特性作了分析,基于对各种算法优势的融合,提出了一种改进的基于模糊形态学的图像分割算法。仿真对比研究实验中,用多种算法对图像进行分割处理,结果显示采用所提出的算法其分割效果更好,可较好地消除传统分水岭算法产生的过分割现象,无伪边缘,实现了目标与背景的分离,经处理后的图像边缘连续,图像更加平滑,柔和性更好,完整地保留了图像的细节。研究结果表明,所提出的图像处理算法有其可行性与可用性。Senentation is a ftmdamental problem in the image analysis. Aimed at the puzzle of edge blurting in image recognition process of original laser, the paper explored a sort of new image segroentation algorithm based on fuzzy morphology. In the paper, it dissected respective superiority and disadvantage of 1FF watershed algorithm and immersion simulation watershed algorithm, based on the fusion with fuzzy morphology, proposed a sort of improved image segmentation algorithm. By means of improved algorithm it made the image segmentation simulation and made the comparative research for lots of other algorithm, and the effect of the simulation demonstrated that it could better eliminate the over se^nentation phenomenon, and there would be no pseudo- edges, and it realized the target separated from the backgrotmd, the continuation after image edge processed. And also it could be better in gentle performance, smooth in image, and held image detail perfectly. The research result shows that the proposed algorithm of image processing has its feasibility and applicability.

关 键 词：激光图像处理 模糊形态学 分水岭算法 图像分割 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TN244[自动化与计算机技术—计算机科学与技术]