Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (14): 172-174.

• 图形、图像、模式识别 • Previous Articles     Next Articles

Improved operators to dilate and erode color images based on morphology

YIN Xing-yun1,2,WANG Jun2   

  1. 1.School of Information Science and Technology,University of Science and Technology of China,Hefei 230026,China
    2.Information Technology Department,Huainan Normal University,Huainan,Anhui 232001,China
  • Received:2007-12-28 Revised:2008-03-17 Online:2008-05-11 Published:2008-05-11
  • Contact: YIN Xing-yun


尹星云1,2,王 峻2   

  1. 1.中国科技大学 信息科学技术学院 计算机系,合肥 230026
    2.淮南师范学院 信息技术系,安徽 淮南 232001
  • 通讯作者: 尹星云

Abstract: By analyzing both the strengths and weaknesses of the color space of RGB,LSH,L*a*b*,this paper puts forward a new order of color vector,which focuses on RGB colorspacith a combination of the revised distance order and lexicographical order.Basic operators of color image morphology are established.The result of the experiment demonstrates that the color vector order has good property.The algorithm of color image morphology based on this order can better keep the peripheral information and hue of the image and has the strong point of unnecessity of exchanging between different color spaces.

Key words: color space, vector order, mathematical morphology, dilation, erosion

摘要: 分析了RGB、LSH、L*a*b*颜色空间的优缺点,提出了一种面向RGB颜色空间的、基于改进的距离序和字典序相结合的颜色向量序并分析了其性能;定义了彩色图像形态学的基本算子,实现了彩色图像形态学重构开算法。实验结果表明提出的颜色向量序性能良好,建立在该向量序上的彩色图像形态学算法能够较好的保持图像的边缘信息和色调,并且亮度均匀,与原图像相比对比度均匀增强,重构开算法的实验结果比原文献中方法有了较大的改进;具有不需要在不同的颜色空间之间转换的优点。

关键词: 颜色空间, 向量序, 数学形态学, 膨胀, 腐蚀