Objective: The objective of this study is an Automatic Extraction of road networks from very high resolution satellite images. It is an important research area in remote sensing field.
Methods/Analysis: We present fully automatic road extraction from high resolution satellite images using modified self organizing map. Firstly, it focuses a road detection using self organizing map algorithm. At the end, the T-Cluster method is used to improve the segmentation in road networks.
Findings: Experimental results show the significant accuracy (90%) and efficiency of proposed approach.
Application/Improvement: The modified T-SOM technique provides the resources for readily creating, maintaining and updating the road transportation databases used in vehicle tracking and traffic management.