2dpca-based polarization SAR image classification method

Abstract

The invention discloses a polarization SAR image classification method based on 2DPCA, which mainly solves the problem of low classification accuracy of the existing unsupervised polarization SAR classification method. The implementation steps are as follows: perform Freeman decomposition on each pixel point to extract three kinds of scattering power of the pixel point; divide the image according to the obtained scattering power to obtain three categories; for each category obtained, use 2DKPCA for Adaptive dimensionality reduction classification; finally, iteratively classify the pre-classified image with Wishart classifier to get the final classification result. Compared with the classical classification method, the invention has more rigorous division of the polarimetric SAR image, better classification effect, relatively less computational complexity, and can be used for ground object classification and target recognition of the polarimetric SAR image.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to the application in the field of classification of polarimetric SAR images, in particular to a method for classification of polarimetric SAR images based on 2DPCA, which can be used for classification of polarimetric SAR images and object recognition . Background technique [0002] Polarization SAR radar can obtain richer target information, and has a wide range of research and application values ​​in agriculture, forestry, military, geology, hydrology and oceans, such as the identification of ground object types, crop growth monitoring, yield evaluation, Object classification, sea ice monitoring, land subsidence monitoring, target detection and marine pollution detection, etc. The purpose of polarimetric SAR image classification is to use the polarization measurement data obtained by airborne or spaceborne polarization sensors to determine the category to which each pixel...

AI Technical Summary

Problems solved by technology

There are two defects in the H / α classification: one is that the division of regions is too arbitrary; the other is that when several different features coexist in the same region, they cannot be effectively distinguished

[0004] Lee et al proposed the H / α-Wishart unsupervised classification method based on H / α target decomposition and Wishart classifier, see Lee J S, Grunes M R, Ainsworth T L, et a1.Unsupervised classification using polarimetric decomposition and the complex Wishart classifier[J ].IEEETrans.Geosci.Remote Sensing.1999, 37(5):2249-2258. This method adds Wishart iteration to the original H / α classification, mainly using Wishart classification for the 8 categories after H / α division The detector re-divides each pixel, thus effectively improving the classification accuracy, but there are also shortcomings that cannot maintain the polarization scattering characteristics of various types.

Images