Innovative Solutions Based on the EM-Algorithm for Covariance Structure Detection and Classification in Polarimetric SAR Images

Abstract

This paper addresses the challenge of identifying the polarimetric covariance matrix (PCM) structures associated with a polarimetric SAR image. Interestingly, such information can be used, for instance, to improve the scene interpretation or to enhance the performance of (possibly PCM-based) segmentation algorithms as well as other kinds of methods. To this end, a general framework to solve a multiple hypothesis test is introduced with the aim to detect and classify contextual spatial variations in polarimetric SAR images. Specifically, under the null hypothesis, only one unknown structure is assumed for data belonging to a 2-dimensional spatial sliding window, whereas under each alternative hypothesis, data are partitioned into subsets sharing different PCM structures. The problem of partition estimation is solved by resorting to hidden random variables representative of covariance structure classes and the expectation maximization (EM) algorithm. The effectiveness of the proposed detection strategies is demonstrated on both simulated and real polarimetric SAR data also in comparison with existing classification algorithms.

Synthetic aperture radar remote sensing scene classification based on fuzzy co-occurrence networks

“…However, in SAR image classification tasks, learning effective representations to utilize discriminative information of different land cover types plays a crucial role compared to classifier model training. 5 Extraction of discriminant characteristics and the development of effective classifiers for SAR remote sensing scene categorization have made significant progress in the last several decades. The two primary types of existing approaches are deep learning methods and classical machine learning algorithms.…”

“…For most existing SAR image classification systems, their performance is largely dependent on significant feature description and high-quality labeled samples to train a robust classifier. However, in SAR image classification tasks, learning effective representations to utilize discriminative information of different land cover types plays a crucial role compared to classifier model training 5 …”

Accuracy Improvement of High-Resolution Wide-Swath Spaceborne Synthetic Aperture Radar Imaging with Low Pule Repetition Frequency

“…Spaceborne synthetic aperture radar (SAR) can actively achieve high-resolution, wideswath, all-day, and all-weather imaging of the observation area [1][2][3][4][5][6][7]. Therefore, it has been widely used in military reconnaissance, mapping, resource surveys, and other fields.…”