Polarized SAR Image Classification Based on Deep Adaptive Ridgelet Network

Abstract

The invention discloses a polarization SAR image classification method based on a deep adaptive ridge wave network. The implementation steps of the method are: (1) filtering the polarization SAR image to be classified; (2) extracting the filtered polarization SAR Image features; (3) Feature combination and normalization; (4) Training Restricted Boltzmann Machine RBM; (5) Training adaptive ridgelet network; (6) Adjusting adaptive ridgelet network parameters; (7) ) image classification. Compared with the existing methods, the present invention has a stronger ability to express polarimetric SAR image features, can better learn higher-level features from original low-level high-dimensional features, and the classifier has a more flexible structure and fast parallelism Processing speed and strong fault tolerance and robustness. The invention can effectively improve the classification accuracy of the polarimetric SAR image, reduce the computational complexity, have better denoising effect and improve the image quality at the same time.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a polarization SAR image classification based on a deep self-adaptive ridge wave network, which can be used for the classification and recognition of the polarization SAR image, and can effectively improve the accuracy of the polarization SAR image classification. Background technique [0002] Polarimetric Synthetic Aperture Radar (SAR) has become one of the most advanced sensors in the field of remote sensing, and polarimetric SAR image classification is an important research technology for SAR image interpretation. Polarimetric SAR (Polarimetric SAR, PolSAR) can describe the target more comprehensively, and its measurement data contains more abundant target information, so polarimetric SAR has very obvious advantages in target detection, classification and parameter inversion. The purpose of polarimetric SAR image classification is to use the polarization m...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to address the above-mentioned deficiencies in the prior art, to propose a polarization SAR image classification method based on a deep adaptive ridge wave network, to adaptively search for and fully excavate the polarization and direction characteristics of the polarization SAR data, and to Overcome the shortcomings of poor adaptability of traditional methods, while improving the classification accuracy of polarimetric SAR images and reducing time complexity

Images