Detection and Extraction of Roads using Cartosat-2 High Resolution Satellite Imagery

The extraction of roads from panchromatic images has many insightful applications in the fields of urban planning, setting up of transportation, disaster management and cartography in geographical information systems (GIS). The process of extracting roads from high-resolution images is composite, due to the presence of different noises (i.e., buildings, shadows, clouds etc.). Various image processing techniques and various quality measures are applied on the high-resolution remote sensing satellite images to improve the quality of the image and interactively extract the information of roads. Cartosat-2 images available in Bhuvan website of ISRO are taken for testing the validity of proposed method. The proposed method enhances the images using Contrast Limited Adaptive Histogram Equalization (CLAHE) and Line Detector for detecting road segments. Connected component analysis (CCA) is performed on segmented image for connection of disconnected objects in the segmented image. Morphological operations fill the holes caused by the presence of shadows, buildings and trees on the road surface.