A Lake Extraction Method Combining the Object-Oriented Method with Boundary Recognition

The China–Pakistan Economic Corridor is the pilot area of Belt and Road, where glaciers lakes are widely distributed. Recent years, global warming has accelerated expansion glacier lakes, which increased risk natural disasters such as lake outburst. It important to monitor in this region. In paper, we propose a method combining object-oriented image analysis with boundary recognition (OOBR) extract several study areas (CPEC). This recognized symmetrical characteristic according principle seed growth watershed algorithm, can correct extracted by method. overall accuracy proposed up 98.5% Landsat series images. experiments also show that our always higher than different segmentation scales mentioned paper. improved on basis results obtained method, more robust seeds correction algorithm. Therefore, obtain high extraction while reducing complexity extraction.