A method for water extraction from remote sensing images

Abstract

The invention relates to a method for extracting water bodies from remote sensing images, which integrates a variety of algorithms, including image search algorithms, fuzzy clustering algorithms, convex hull point counterclockwise sorting algorithms, and symbolic distance function initialization algorithms, and builds a new control strategy that can For the water body in the remote sensing image, it is more convenient and accurate to extract, and according to the CIE LAB, HSL color model and Gaussian normalized water index GNDWI, the feature space is constructed for the remote sensing image, which effectively solves the application of the level set algorithm to multiple targets The convergence speed is slow, or even unable to converge; not only that, but also the parallel processing method is introduced, and the iterative convergence operation of the level set function is performed on multiple polygonal target water bodies at the same time, which greatly improves the work efficiency of the entire design technical solution.

Description

technical field [0001] The invention relates to a method for extracting a water body from a remote sensing image, and belongs to the technical field of remote sensing image processing. Background technique [0002] With the advancement of computer science and space science and the rapid development of sensor technology and communication technology, remote sensing technology has also developed rapidly, and the resolution of remote sensing images has been greatly improved: the spatial resolution has been increased from the initial 80 meters to the present Even the resolution of military remote sensing data has broken through the meter level; the spectral resolution has also reached the nanometer level, and for hyperspectral and hyperspectral data, it can be subdivided into thousands of bands at this stage; In terms of time resolution, the satellite revisit cycle has been shortened from dozens of days to a few days now. [0003] The development of remote sensing technology has...

AI Technical Summary

Problems solved by technology

[0025] Researchers have proposed a variety of level set algorithms, all of which evolve from a single target contour inward or outward. This algorithm has a slow convergence speed when dealing with images with multiple targets, and encounters grayscale Inhomogeneous conditions can also cause the level set function to oscillate and not converge

Images