Automatic Extraction of Roads from High Resolution Aerial and Satellite Images with Heavy Noise
Aerial and satellite images are information rich. They are also complex to analyze. For GIS systems, many features require fast and reliable extraction of roads and intersections. In this paper, we study efficient and reliable automatic extraction algorithms to address some difficult issues that are commonly seen in high resolution aerial and satellite images, nonetheless not well addressed in existing solutions, such as blurring, broken or missing road boundaries, lack of road profiles, heavy shadows, and interfering surrounding objects. The new scheme is based on a new method, namely reference circle, to properly identify the pixels that belong to the same road and use this information to recover the whole road network. This feature is invariable to the shape and direction of roads and tolerates heavy noise and disturbances. Road extraction based on reference circles is much more noise tolerant and flexible than the previous edge-detection based algorithms. The scheme is able to extract roads reliably from images with complex contents and heavy obstructions, such as the high resolution aerial/satellite images available from Google maps.
[1] Y. Li, R. Briggs. Scalable and error tolerant automated georeferencing
under affine transformations. IEEE International Geoscience and
Remote Sensing Symposium, Boston, MA, July 2008.
[2] Y. Li, R. Briggs. Automated georeferencing based on topological point
pattern matching. The International Symposium on Automated
Cartography (AutoCarto), Vancouver, WA, June 2006.
[3] J.B. Mena. State of the Art on Automatic Road Extraction for GIS
Update: a Novel Classification. Pattern Recognition Letters,
24(16):3037-3058, 2003.
[4] M.-F. Auclair-Fortier, D. Ziou, C. Armenakis, and S. Wang. Survey of
Work on Road Extraction in Aerial and Satellite Images. Technical
Report 241, Département de mathématiques et d-informatique,
Université de Sherbrooke, 1999.
[5] G. Vosselman, J.D. Knecht. Road Tracking by Profile Matching and
Kalman Filtering. Workshop on Automatic Extraction of Man-Made
Objects from Aerial and Space Images, pages 265-274, 1995.
[6] D.M. Mckeown, J.L. Denlinger. Cooperative Methods for Road
Tracking in Aerial Imagery. Workshop Computer Vision Pattern
Recognition, pages 662-672, 1988.
[7] A. Gruen, H. Li. Semi-automatic Linear Feature Extraction by Dynamic
Programming and LSB-Snakes. Photogrammet Eng. Remote Sensing
63, pages 985-995, 1997.
[8] S.B. Hinz, A. Ebner. Modeling Contextual Knowledge for Controlling
Road Extraction in Urban Areas. IEEE/ISPRS Joint Workshop Remote
Sensing Data Fusion over Urban Areas, 2001.
[9] T. Ohlhof, T. Emge, W. Reinhardt, K. Leukert, C. Heipke, K. Pakzad.
Generation and Update of VMAP data using satellite and airbone
imagery. Remote Sensing, vol 33, pages 763-768, 2000.
[10] A. Baumgartner, C.T. Steger, H. Mayer, and W. Eckstein. Multi-
Resolution, Semantic Objects, and Context for Road Extraction. In
Wolfgang Förstner and Lutz Pl├╝mer, editors, Semantic Modeling for the
Acquisition of Topographic Information from Images and Maps, pages
140-156, Basel, Switzerland, 1997. Birkhäuser Verlag.
[11] A. Baumgartner, C.T. Steger, C. Wiedemann, H. Mayer, W. Eckstein,
and H. Ebner. Update of Roads in GIS from Aerial Imagery:
Verification and Multi-Resolution Extraction. Proceedings of
International Archives of Photogrammetry and Remote Sensing, XXXI
B3/III:53-58, 1996.
[12] D. Geman and B. Jedynak. An Active Testing Model for Tracking
Roads in Satellite Images. IEEE Transactions on Pattern Analysis and
Machine Intelligence, 18(1):1-14, January 1996.
[13] E. Christophe, J. Inglada. Robust Road Extraction for High Resolution
Satellite Images. ICIP 2007, pages 437-440, October 2007.
[14] W.J. Rucklidge. Efficient Visual Recongnition Using Hausdorff
Distance. Lecture Notes in Computer Science. Springer-Verlag, 1996.
[15] G. Lisini, C. Tison, F. Tupin, P. Gamba. Feature Fusion to Improve
Road Network Extraction in High Resolution SAR Images. IEEE
Geoscience Remote Sensing Letter, Vol 3, No. 2, pages 217-221, 2006.
[16] X. Jin, C.H. Davie. An Integrated System for Automatic Road Mapping
from High Resolution Multispectural Satellite Imagery by Information
Fusion. Information Fusion. Vol 6(4), pages 257-273, 2005.
[1] Y. Li, R. Briggs. Scalable and error tolerant automated georeferencing
under affine transformations. IEEE International Geoscience and
Remote Sensing Symposium, Boston, MA, July 2008.
[2] Y. Li, R. Briggs. Automated georeferencing based on topological point
pattern matching. The International Symposium on Automated
Cartography (AutoCarto), Vancouver, WA, June 2006.
[3] J.B. Mena. State of the Art on Automatic Road Extraction for GIS
Update: a Novel Classification. Pattern Recognition Letters,
24(16):3037-3058, 2003.
[4] M.-F. Auclair-Fortier, D. Ziou, C. Armenakis, and S. Wang. Survey of
Work on Road Extraction in Aerial and Satellite Images. Technical
Report 241, Département de mathématiques et d-informatique,
Université de Sherbrooke, 1999.
[5] G. Vosselman, J.D. Knecht. Road Tracking by Profile Matching and
Kalman Filtering. Workshop on Automatic Extraction of Man-Made
Objects from Aerial and Space Images, pages 265-274, 1995.
[6] D.M. Mckeown, J.L. Denlinger. Cooperative Methods for Road
Tracking in Aerial Imagery. Workshop Computer Vision Pattern
Recognition, pages 662-672, 1988.
[7] A. Gruen, H. Li. Semi-automatic Linear Feature Extraction by Dynamic
Programming and LSB-Snakes. Photogrammet Eng. Remote Sensing
63, pages 985-995, 1997.
[8] S.B. Hinz, A. Ebner. Modeling Contextual Knowledge for Controlling
Road Extraction in Urban Areas. IEEE/ISPRS Joint Workshop Remote
Sensing Data Fusion over Urban Areas, 2001.
[9] T. Ohlhof, T. Emge, W. Reinhardt, K. Leukert, C. Heipke, K. Pakzad.
Generation and Update of VMAP data using satellite and airbone
imagery. Remote Sensing, vol 33, pages 763-768, 2000.
[10] A. Baumgartner, C.T. Steger, H. Mayer, and W. Eckstein. Multi-
Resolution, Semantic Objects, and Context for Road Extraction. In
Wolfgang Förstner and Lutz Pl├╝mer, editors, Semantic Modeling for the
Acquisition of Topographic Information from Images and Maps, pages
140-156, Basel, Switzerland, 1997. Birkhäuser Verlag.
[11] A. Baumgartner, C.T. Steger, C. Wiedemann, H. Mayer, W. Eckstein,
and H. Ebner. Update of Roads in GIS from Aerial Imagery:
Verification and Multi-Resolution Extraction. Proceedings of
International Archives of Photogrammetry and Remote Sensing, XXXI
B3/III:53-58, 1996.
[12] D. Geman and B. Jedynak. An Active Testing Model for Tracking
Roads in Satellite Images. IEEE Transactions on Pattern Analysis and
Machine Intelligence, 18(1):1-14, January 1996.
[13] E. Christophe, J. Inglada. Robust Road Extraction for High Resolution
Satellite Images. ICIP 2007, pages 437-440, October 2007.
[14] W.J. Rucklidge. Efficient Visual Recongnition Using Hausdorff
Distance. Lecture Notes in Computer Science. Springer-Verlag, 1996.
[15] G. Lisini, C. Tison, F. Tupin, P. Gamba. Feature Fusion to Improve
Road Network Extraction in High Resolution SAR Images. IEEE
Geoscience Remote Sensing Letter, Vol 3, No. 2, pages 217-221, 2006.
[16] X. Jin, C.H. Davie. An Integrated System for Automatic Road Mapping
from High Resolution Multispectural Satellite Imagery by Information
Fusion. Information Fusion. Vol 6(4), pages 257-273, 2005.
@article{"International Journal of Information, Control and Computer Sciences:51064", author = "Yan Li and Ronald Briggs", title = "Automatic Extraction of Roads from High Resolution Aerial and Satellite Images with Heavy Noise", abstract = "Aerial and satellite images are information rich. They are also complex to analyze. For GIS systems, many features require fast and reliable extraction of roads and intersections. In this paper, we study efficient and reliable automatic extraction algorithms to address some difficult issues that are commonly seen in high resolution aerial and satellite images, nonetheless not well addressed in existing solutions, such as blurring, broken or missing road boundaries, lack of road profiles, heavy shadows, and interfering surrounding objects. The new scheme is based on a new method, namely reference circle, to properly identify the pixels that belong to the same road and use this information to recover the whole road network. This feature is invariable to the shape and direction of roads and tolerates heavy noise and disturbances. Road extraction based on reference circles is much more noise tolerant and flexible than the previous edge-detection based algorithms. The scheme is able to extract roads reliably from images with complex contents and heavy obstructions, such as the high resolution aerial/satellite images available from Google maps.
", keywords = "Automatic road extraction, Image processing, Feature extraction, GIS update, Remote sensing, Geo-referencing", volume = "3", number = "6", pages = "1513-7", }
