Title :
Multiple level set region based single line road extraction
Author :
Akram, Khondekar Mahabub ; Lutfe Elahi, Mirza M. ; Amin, M. Ashraful
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., North South Univ., Dhaka, Bangladesh
Abstract :
A new region detection approach for GIS system has been proposed to detect the roads in image. After extraction of road parts from the other parts of image, binarized road features are mined and then used to get a single line road features by thinning. The pruning of the extra unrelated branches and linking of the related branches have been done using Bottom-Top Nega-Max Formulation and Gradient Circum-circle Based Linking. Our proposed method shows high accuracy in detecting roads from satellite images. We have proved by experiment and analysis that proposed method of road extraction is best suited to deal with an average of 95.6% accuracy.
Keywords :
artificial satellites; feature extraction; geographic information systems; geophysical image processing; gradient methods; image thinning; object detection; GIS system; binarized road feature mining; bottom-top nega-max formulation; branch pruning; gradient circum circle based linking; image thinning; multiple level set region; region detection approach; road detection; satellite images; single line road extraction; Abstracts; Accuracy; Image segmentation; Minimization; Alpha-beta pruning; Gradient linking; Level set; Linking; Multi-channel; Pruning; Statistical minimization;
Conference_Titel :
Machine Learning and Cybernetics (ICMLC), 2013 International Conference on
Conference_Location :
Tianjin
DOI :
10.1109/ICMLC.2013.6890773
