期刊文献+

台风螺旋云带骨架跟踪方法的研究 被引量:9

The Study on Automatic Tracking Method of Typhoon Spiral Cloud Bands
在线阅读 下载PDF
导出
摘要 用紧邻交叉点的已跟踪编码过的一段象素点的编码信息,用最小二乘法向前作线性预测,求出交叉点处下一步的预测搜索方向,当用线性预测方法无法求出预测方向时,则计算交叉点前的已跟踪编码过的一段象素点的平均链码值作为预测搜索方向,最后以交叉点的各邻点方向中与预测搜索方向最接近的方向作为下一点的搜索方向。该方法有效地克服了基于链码的跟踪在交叉点处容易发生迷向的现象,同时跟踪速度和自动化程度得到了较大的提高。 In this paper, an automatic skeleton tracking method is proposed according to the linear predicting technique based on chain code and the concept of average chain code. In this method, when encountering an intersect during tracking, firstly, linear predicting technique is used to calculate the track direction at intersects based on chain code sequence of a section of tracked pixels adjacent to the intersect. When the linear predicting is an insolvable problem, the average chain code of pixels tracked can be calculated as predicted direction. Then refer to predicted direction, the nearest direction of all neighboring directions to predicted direction at intersect can be used as the tracking direction at intersect. During the course of tracking the skeleton of typhoon spiral cloud bands, this method effectively overcome the defect of direction lost at intersect based on chain code, and the speed and automation of tracking is also improved.
出处 《计算机工程》 CAS CSCD 北大核心 2001年第10期152-154,共3页 Computer Engineering
关键词 FREEMAN链码 线性预测 骨架跟踪 台风 自动定位 螺旋云带 图象处理 Freeman chain code Linear predicting Average chain code Skeleton tracking
  • 相关文献

参考文献3

二级参考文献10

  • 1谭建荣,彭群生.基于图形约束的工程图扫描图象直线整体识别方法[J].计算机学报,1994,17(8):561-569. 被引量:27
  • 2陆宗骐.图象处理领域轮廓跟踪及应用[J].中国计算机用户,1994(10):49-52. 被引量:19
  • 3吴健康.数学图象分析[M].北京:人民邮电出版社,1989..
  • 4Hopfield J J. Neural Networks and physical systems with emergent collective computational abilities , Proc. Natl. Acad, Sci.USA. 79, 1982:2554-2558.
  • 5Hopfield J J,Tank D W. Computing with neural circuits: a model. Science 233, 1986:625-633.
  • 6Kasturi R,Shih R,Flecher L. An approach for automatic recognition of graphics ,8th PRIP, Paris ,877-879.
  • 7Kasturi R ,Alemany J. Information extraction from images of paper-based maps. IEEE Trans. Software Eegineering (Image Databases ), 14:671-675.
  • 8Ejiri M. Automatic recognition of design drawings and maps. 7th.Intl. Conf. on Pattern Recognition, Montreal. Canada, 1984:1296-1305.
  • 9Hopfield J J. Neurons with graded response have collective computational properties like those of two-state neurons. Proe. Natl.Acad. Sci. USA 81,1984:3088-3092.
  • 10孙贞寿,钱刚,黄琨.基于Hopfield神经网络的交叉线识别算法[J].中国图象图形学报(A辑),1998,3(8):684-687. 被引量:5

共引文献25

同被引文献62

引证文献9

二级引证文献30

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