Road Extraction Based on Object-Oriented from High-Resolution Remote Sensing Images

Because of the precision of the information extraction is lack. The paper is using the data source with quick-bird, processing the segmentation of high-resolution remote sensing images through the watershed algorithm of controlling marked, controlling the over-segmentation of watershed algorithm with the self-adaption space filter algorithm by matlab to attain a certain precision, to meet the need of land monitoring. To avoid the noise of spectra, we use the geometry properties, texture properties and so on to extract the features. Comparing with other classification, the object-oriented method is fast, high precision and high noise resisting. The research is important for land monitoring and GIS database updating and the rapid reaction.