INDUCTIVE CLUSTERING: AUTOMATING LOW-LEVEL SEGMENTATION IN HIGH RESOLUTION IMAGES *

In this paper we present a new classification technique for segmenting remotely sensed images, based on cluster analysis and machine learning. Traditional segmentation techniques which use clustering require human interaction to fine -tune the clustering algorithm parameters and select good clusters. Our technique applies inductive learning techniques using C4.5 to learn the parameters and pick good clusters automatically. The techniques are demonstrated on level 1 of RAIL, a hierarchical road recognition system we have developed.

[1]  Arcot Sowmya,et al.  JUNCTION EXTRACTION FROM HIGH RESOLUTION IMAGES BY COMPOSITE LEARNING , 2000 .

[2]  W. Harvey.  Performance evaluation for road extraction , 1999 .

[3]  Arcot Sowmya,et al.  Recent Progress on RAIL : Automating Clustering and Comparison of Different Road Classification Techniques on High Resolution Remotely Sensed Imagery , 2002 .

[4]  C. Steger,et al.  AUTOMATIC ROAD EXTRACTION BASED ON MULTI-SCALE, GROUPING, AND CONTEXT , 1999 .

[5]  Salesh Singh Arcot Sowmya,et al.  RAIL : Road Recognition from Aerial Images Using Inductive Learning , 1998 .

[6]  金田 重郎,et al.  C4.5: Programs for Machine Learning (書評) , 1995 .

[7]  Jacky Desachy,et al.  Remote Sensed Image Processing and Geographic Features , 1999 .

[8]  Ivan Laptev,et al.  Automatic extraction of roads from aerial images based on scale space and snakes , 2000, Machine Vision and Applications.

[9]  David G. Lowe,et al.  Vista: a software environment for computer vision research , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[10]  Yandong Wang,et al.  Automatic Road Extraction from Aerial Images , 1998, Digit. Signal Process..

[11]  Sholom M. Weiss,et al.  Computer Systems That Learn , 1990 .

[12]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[13]  Ricardo Barandela,et al.  Ongoing learning for supervised pattern recognition , 2001, Proceedings XIV Brazilian Symposium on Computer Graphics and Image Processing.