Fully automatic road network extraction from satellite images

In this paper a fully automatic road detection algorithm is introduced. It comprises of pre-processing the image via a series of wavelet based filter banks and reducing the yielding data into a single image which is of the same size as the original optical grayscale satellite image, then utilizing a fuzzy inference algorithm to carry out the road detection which can then be used as an input to a geographical information system for cartographic or for other purposes that are in need. We use a trous algorithm twice with two different wavelet bases in order to filter and de-noise the satellite image. Each wavelet function resolves features at a different resolution level associated with the frequency response of the corresponding FIR filter. Resulting two images are fused together using Karhounen-Louve transform (KLT) which is based on principal component analysis (PCA). This process underlines the prominent features of the original image as well as de-noising it, since the prominent features appear in both of the wavelet transformed images while noise does not strongly correlate between scales. Next a fuzzy logic inference algorithm which is based on statistical information and on geometry is used to extract the road pixels.