In this research roads extraction / mapping at a sub-pixel and pixel levels was evaluated in the northern part of Sinai of Egypt. Three different image fusion techniques have been used to merge two co-registered data, microwave (Radarsat) and optical (Egyptsat), to improve the classification accuracy. Spatial, spectral, and radiometric qualities of the fused images have been evaluated.Three different fusion techniques are used in the current research, namely: Intensity-Hue-Saturation (IHS), Hue-Saturation-Value (HSV) and Wavelet fusion.The resulted three fused images, combining the Egyptsat-1 and Radarsat- 2 images, were fed to four different classification methods, namely: fuzzy, neural network (as a soft classifiers), the maximum likelihood, and minimum distance (as a hard classifiers), where the classification delineated five classes, namely: water, urban, roads, sand, and agriculture. Then, separate roads class only. the results prove that the approach is excellent for extracting roads from medium resolution images.
Rana Rezk Mahmoud Mahmoud Omran
Egyptian Civil Engineer,geomatics&projects Engineer(Photogrammetry&flying Department)technical Office Engineer,junior Design Engineer.and I was graduated from HELWAN University 2011 with the degree of Science Civil Engineering plus degree honer.Master of Science 2016: Survey Engineering(Road Detection Using Multi Resolution Remote)
Number of Pages:
LAP LAMBERT Academic Publishing
civil engineering, Remote sensing, surveying engineering, Surveying and Photogrammetry, Roads and Transportation
LAW / General