Classification Method of Polarized SAR Image Based on Pyramid Sampling and Support Vector Machine

Abstract

The invention discloses a polarization SAR image classification method based on pyramid sampling and SVM, which mainly solves the problem of low classification accuracy in the prior art. The implementation steps are as follows: firstly, filter the polarimetric SAR image; secondly, extract the sampling scattering feature of the polarimetric SAR image based on pyramid sampling; finally, extract the polarimetric scattering feature and wavelet texture feature of the polarimetric SAR image, and convert the sampling scattering feature, polarization scattering feature and wavelet texture feature are combined to obtain the combined feature, and the combined feature is used to train the support vector machine classifier, and the trained classifier is used to classify the polarimetric SAR image, and the classified polarimetric SAR image Do the coloring. The invention has better denoising effect, improves image quality and classification accuracy, and can be used for target recognition of polarimetric SAR images.

Description

technical field [0001] The invention belongs to the field of radar technology, in particular to image classification of polarization synthetic aperture radar SAR, which can be used for target recognition. Background technique [0002] Polarization synthetic aperture radar (SAR) has become one of the important development directions of synthetic aperture radar at home and abroad. Compared with single-polarization radar images, polarization SAR images can provide more ground object information. Image classification is one of the important contents of polarimetric SAR image interpretation, which has been widely used in military and civilian fields. Fast and accurate SAR image classification is the prerequisite for various practical applications. Therefore, it is of great significance to study the classification of polarimetric SAR images. Classification methods have always been a hot spot in frontier research in this field. Many polarimetric SAR image classification methods h...

AI Technical Summary

Problems solved by technology

At present, the algorithms involved in polarimetric SAR image classification include: traditional image processing algorithms, representative algorithms include mean clustering algorithm, ISODATA algorithm, watershed algorithm, graph theory method, etc. Although these methods are based on theoretically mature classifiers, but Does not make full use of the target scattering mechanism for polarization SAR image classification; based on the classification method of the general pyramid model, the advantage of the pyramid model is that the domain information is taken into account. In the marked area, the sample division is very clean, but this method also has boundary division Not obvious, and in the case of few sample points, the accuracy is not high

Images