A method for detecting changes in land cover types based on time series polsar images

Abstract

The present invention provides a land cover type change detection method based on time series PolSAR images. The purpose is to solve the problem that the current change detection method is difficult to make full use of the time dimension information between time series images, and usually requires a large number of high-quality training for each image. sample etc. The invention uses the Omnibus hypothesis test likelihood ratio algorithm to fully mine the time dimension information of the historically accumulated time series PolSAR images, and combines the marked category labels, classification thematic maps and other rich prior knowledge, and uses active learning and associated knowledge transfer learning And other machine learning algorithms to automatically label the category labels of each scene image, obtain reliable training samples, and then realize high-precision extraction of dynamic change information of land cover categories.

Description

technical field [0001] The invention belongs to the field of polarimetric synthetic aperture radar image interpretation, and is a new method for dynamically changing land cover types by using time-series polarimetric SAR (Polarimetric Synthetic Aperture Radar, PolSAR) images. Background technique [0002] With the continuous and rapid development of the global economy and the transformation of the natural environment by humans, the types of surface land cover are changing frequently. Among them, land cover change detection based on remote sensing images is an important means to study regional and even global land cover type changes. Because optical remote sensing images have the advantages of wide coverage, short revisit period, and easy image acquisition, land cover change detection methods based on remote sensing images have been widely used in various fields. However, optical remote sensing sensors are easily affected by factors such as weather and light, which greatly l...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problems that the land cover supervision change detection method based on time series PolSAR images is difficult to make full use of the time dimension information between time series images, and usually requires a large number of high-quality training samples for each image.

Images