Classification Method of Polarization SAR Ground Objects Based on Depth RPCA

Abstract

The invention discloses a polarimetric SAR ground object classification method based on deep RPCA. The implementation steps of the present invention are: (1) reading in polarimetric SAR images; (2) filtering; (3) extracting features; (4) normalizing feature groups; (5) selecting training samples and test samples; (6) training (7) training the second layer of deep robust principal component analysis RPCA; (8) training support vector machine SVM; (9) generating superpixels; (10) classification; (11) Calculate the classification accuracy; (12) output the result. Compared with the scattering features of polarimetric SAR images, the image features extracted by the present invention contain more abundant ground object information, and when used for classification, the classification accuracy is effectively improved, and can be used for target detection and target detection in polarimetric SAR images. identify.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a deep RPCA-based polarization SAR object classification method in the technical field of image classification. This method can be applied to target detection and target recognition of polarimetric SAR images, and can effectively improve the classification accuracy of polarimetric SAR images. Background technique [0002] Polarimetric SAR images describe the observed land cover and targets by transmitting and receiving polarized radar waves, and it is one of the most advanced sensors in the field of remote sensing in recent years. As an important means of remote sensing image acquisition, polarimetric SAR images are widely used in agriculture, forestry, military affairs, oceanography, hydrology and geology. The purpose of polarimetric SAR image classification is to use the polarimetric measurement data obtained by airborne or spaceborne polarimetric sensors to det...

AI Technical Summary

Problems solved by technology

However, the shortcomings of this method are that the extracted scattering, polarization and texture information are simply stacked, and then directly input into the support vector machine (SVM) for classification, which leads to more redundant features in the input features. The remaining information cannot effectively reflect the essential characteristics of the polarimetric SAR image, which greatly reduces the classification efficiency

Images