Automatic Road Extraction of Satellite Imagery using Oscillatory Network

Abstract

Updating road maps is a time consuming operation when performed manually. Unfortunately, no existing methods allow complete and reliable automatiziation of this process. This investigates automatic road extraction from satellite imagery based on oscillatory networks called LEGION. By using this technique we are able to track the road from satellite images very accurately .This is very much useful in many applications such as military purposes, position determination of path sensing in remote areas. The task is divided into three sections. They are image segmentation, medial axis transform and grouping. The first step is image segmentation by LEGION. In the second step, the medial axis of each segment is computed, and the medial axis points corresponding to narrow regions are selected. And the last is the road grouping stage. The proposed method can effectively carry out road extraction from different satellite images for both synthetic and real images.