期刊文献+

结合颜色特征和结构特征的彩色地图分层算法 被引量:2

Automatic Segmentation of Color Map Images Based on Color and Structure Features
下载PDF
导出
摘要 在分析了彩色地图图像分层的特点、难点的基础上 ,提出直接基于RGB颜色模型的分层算法 同时针对分层得到的单版图存在信息丢失、难以恢复的问题 ,提出先对所有目标细化 ,然后再分层的新思路 ,并结合结构特征恢复丢失信息、改善分层效果 最后 。 An algorithm directly based on RGB color model to enhance the quality and performance of color map segmentation is proposed The problem of information losing in segmentation is solved by carrying out segmentation after thinning of all objects and making use of structure features to improve segmentation results Experiments on many real maps illustrate the effectiveness of the above mentioned approach
出处 《计算机辅助设计与图形学学报》 EI CSCD 北大核心 2003年第12期1463-1467,共5页 Journal of Computer-Aided Design & Computer Graphics
关键词 颜色特征 结构特征 彩色地图分层算法 地理信息系统 彩色图像 color map vectorization segmentation color model
  • 相关文献

参考文献8

  • 1Zhang Y J. A survey of evaluation methods for image segmentation [J]. Pattern Recognition, 1996, 29(8): 1335~1346
  • 2Level Otsu N. A threshold selection method from gray-level histogram [J]. IEEE Transactions on Systems, Man and Cybernetics, 1979, 9(1): 62~66
  • 3Brink A D. Thresholding of digital images using two-dimensional entropies [J]. Pattern Recognition, 1992, 25(8): 803~808
  • 4Nalwa V S, Binford T O. On detecting edges [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(6): 248~270
  • 5Azriel Rosenfeld, Thurston M, Lee Y. Edge and curve detection: Further experiments [J]. IEEE Transactions on Computer, 1972, 21(7): 677~710
  • 6Wang J P. Stochastic relaxation on partitions with connected components and its application to image segmentation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1998, 20(8): 619~638
  • 7Gupta J, Sortrakal T. A Gaussian-mixture based image segmentation algorithm [J]. Pattern Recognition, 1998, 31(3): 315~326
  • 8严玉龙,王厚枢.快速并行细化算法[J].南京航空学院学报,1991,23(1):135-140. 被引量:4

共引文献3

同被引文献26

引证文献2

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部