An index based road feature extraction from LANDSAT-8 OLI images

Road feature extraction from the remote sensing images is an arduous task and has a significant role in various applications of urban planning, updating maps, traffic management, etc. In this paper, new band combination (B652) to form road index (RI) OLI multispectral bands based on spectral reflectance asphalt, presented for extraction. The B652 converted by normalization. morphological operators (top-hat or bottom-hat) uses RI enhance roads. To sharpen edges better discrimination features, shock square filter (SSF), proposed. Then, iterative adaptive threshold (IAT) online search with variational min-max Markov random fields (MRF) model are used SSF image segment roads non-roads. extracting using rules connected component analysis. IAT MRF segmentation methods prove proposed able extract features productively. methodology saturation thresholding morphology (SATM), (SMRF), applied several cities India, producing satisfactory results. experimental results quantitative analysis paper.