Road Map Extraction from Satellite Images Using Connected Component Approach

Abstract: The satellite images usually consist of complexity due to presence of several manmade and natural features. Automatic extraction of features from satellite images is a significant analysis for getting an efficient geographical database system and modifying the remote sensing and geographic information system (GIS) areas. We have presented a methodology using image processing and connected component approach for automatic extraction of road network from any high resolution satellite image. Its quite challenging in the areas having complex scenes. There is an algorithm to extract the manmade features from the image and filtering operations are used to correctly extract out the respective feature i.e. the road network from the image. The proposed method follows various steps consisting of Enhancement, Filtering operations, Segmentation etc. The extracted road network can be used for the map creation which shortens the human efforts in creating large scale maps and also it is very efficient in time and cost. The method is simplistic to implement in real time also and can be applied for the real time development in the field of remote sensing and Geographic Information System including rural and urban development.

Keywords: Connected component, Enhancement, Feature extraction, Filtering operations, Road network detection.