基于柔性形态学的抗噪彩色图像边缘检测.PDF

兰州大学学报: 自然科学版, 2016, 52(1) / 2 月 Journal of Lanzhou University ：Natural Sciences，2016 ，52(1) / February 基于柔性形态学的抗噪彩色图像边缘检测 刘燕妮, 张贵仓, 安 静 西北师范大学 数学与统计学院, 兰州 730070 摘 要：针对由RGB 颜色空间各分量高相关性引起的利用分量边缘检测不能有效滤除噪声以及边 缘模糊较粗的问题, 选取彩色图像处理复杂性较低并且符合人眼色彩感受的HSI 颜色空间, 提出一 种将柔性形态学用于彩色图像边缘检测的方法. 利用小尺度算子改进彩色柔性形态学算子的抗噪 性, 即先对H 、S、I 3 个分量使用改进的柔性形态学算子检测边缘, 然后将3 个分量边缘加权得到融合 的彩色图像边缘. 结果表明, 此方法能够在充分保留彩色图像边缘信息的基础上, 有效地抑制噪声, 得到连续、完整、清晰的边缘. 关键词：柔性形态学; 边缘检测; 抗噪; 彩色图像 中图分类号：TP391 文献标识码：A 文章编号：0455-2059(2016)01-0135-07 DOI ：10.13885/j.issn.0455-2059.2016.01.023 Noise-resistance in color image edge detection based on flexible morphology Liu Yan-ni, Zhang Gui-cang, An Jing School of Mathematics and Statistics, Northwest Normal University, Lanzhou 730070, China Abstract: In view of the fact that a high RGB color space correlation between the components used in edge de- tection can effectively filter out noise and blurred edge. A color image processing of lower complexity was cho- sen that could match the human eye color feeling for HSI color space, and a flexible morphology was put for- ward for color image edge detection. A small-scale operator was used to improve the color noise resistance of a flexible morphological operator, the first three components of H, S, I were used to improve the soft morphologi- cal operator edge detection, and then the fused edge was weighted by the three components of the color image edge. The experiment results showed that this method could fully keep the edge information of a color image, thus being able to effectively suppress noise and obtain a continuous complete clear edge. Key words: flexible morphology; edge detection; noise-resistant; color image 图像是人们对客观世界的一种描述和记录. 础, 获取边缘变得愈发重要. 目前,