DocumentCode :
2147004
Title :
Automatic extraction of contour lines from scanned topographic map
Author :
Du Jinyang ; Yumei, Zhang
Author_Institution :
Inst. of Remote Sensing Applications, Chinese Acad. of Sci., Beijing, China
Volume :
5
fYear :
2004
fDate :
20-24 Sept. 2004
Firstpage :
2886
Abstract :
The automatic extraction of contour lines from scanned topographic map is one of the challenging subjects in GIS. Most of the methods available rely heavily on image processing on pixel scale and pay little attention to spatial relationships, which are inherent characteristics of contour lines. This work brings forward a new scheme aiming at the effective realization of automatic extraction of contour lines. Based on mathematical morphology, this scheme tries to acquire spatial relationships of contour lines and make use of them for matching and connecting broken contour lines. The whole process of automatic extraction of contour lines includes the following aspects: acquisition of contour line image by color separation, erasion of image noise, image skeletonizing process, line tracking process, acquisition of spatial relationships by dilation of contour lines, matching and connecting broken contour lines based on spatial relationships, compression and output of vector data. Software was designed according to the scheme and experiments on scanned maps further proved the effectiveness of the method.
Keywords :
edge detection; feature extraction; geographic information systems; image colour analysis; image thinning; mathematical morphology; terrain mapping; topography (Earth); automatic contour line extraction; color separation; contour line dilation; contour line image; contour line matching; data compression; image noise; image processing; image skeletonizing process; line tracking process; mathematical morphology; pixel scale; spatial relationships; topographic map; vector data; Color; Colored noise; Data mining; Geographic Information Systems; Image coding; Image processing; Joining processes; Morphology; Pixel; Software design;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2004. IGARSS '04. Proceedings. 2004 IEEE International
Print_ISBN :
0-7803-8742-2
Type :
conf
DOI :
10.1109/IGARSS.2004.1370296
Filename :
1370296
Link To Document :
بازگشت