Abstract
Algorithms are considered that solve the problem of detecting structural changes in multitemporal satellite images of the Earth's surface. A computational method is formulated for comparing such algorithms; to this end, a special mathematical model of structural changes is introduced that allows one to maximally approximate an experiment to real problems. Optimal parameters of operation of the algorithms are determined by this method, and the results of comparison of the performance of the algorithms are presented. Experiments on real data show that the algorithms proposed are suitable for practical application.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have