A Novel Relational-Based Transductive Transfer Learning Method for PolSAR Images via Time-Series Clustering

Abstract

The combination of transfer learning and remote sensing image processing technology can effectively improve the automation level of image information extraction from a remote sensing time series. However, in the processing of polarimetric synthetic aperture radar (PolSAR) time-series images, the existing transfer learning methods often cannot make full use of the time-series information of the images, relying too much on the labeled samples in the target domain. Furthermore, the speckle noise inherent in synthetic aperture radar (SAR) imagery aggravates the difficulty of the manual selection of labeled samples, so these methods have difficulty in meeting the processing requirements of large data volumes and high efficiency. In lieu of these problems and the spatio-temporal relational knowledge of objects in time-series images, this paper introduces the theory of time-series clustering and proposes a new three-phase time-series clustering algorithm. Due to the full use of the inherent characteristics of the PolSAR images, this algorithm can accurately transfer the labels of the source domain samples to those samples that have not changed in the whole time series without relying on the target domain labeled samples, so as to realize transductive sample label transfer for PolSAR time-series images. Experiments were carried out using three different sets of PolSAR time-series images and the proposed method was compared with two of the existing methods. The experimental results showed that the transfer precision of the proposed method reaches a high level with different data and different objects and it performs significantly better than the existing methods. With strong reliability and practicability, the proposed method can provide a new solution for the rapid information extraction of remote sensing image time series.

Convolutional neural networks for global human settlements mapping from Sentinel-2 satellite imagery

“…Transfer learning is a paradigm in DL to solve a target problem by reusing the learning with minor modifications from a different but related source problem. Qin et al, [74] review transfer learning in remote sensing applications and categorize the methods into four families depending on what is being transferred:  instance-based transfer which uses partial training samples in the source domain to improve the performance of the model of the target domain [75];  feature representation-based transfer [76] which assists the target domain classifier to learn a more effective feature expression from the source domain and improve its performance;  relational knowledge transfer [77] where knowledge among the data in the source domain is transferred to the target domain;…”

Transfer Learning: A Paradigm for Machine Assisted Knowledge Transfer

“…The authors in (2) categorize transfer learning as inductive, transductive, and unsupervised based on the availability/non-availability of labeled source and/or target data. (12) implements transductive transfer learning in remote sensing image processing. It uses the labels of the source domain images directly onto the target domain images because the time-series samples do not change over time.…”

Eigenvalue Statistical Components-Based PU-Learning for PolSAR Built-Up Areas Extraction and Cross-Domain Analysis

“…To reduce the domain shifts between source and target domains, classifier-level DA, invariant feature selection DA and adapting feature distribution representation methods were widely studied. Among them, the adapting feature distribution representation strategy has attracted more and more attention in cross-domain RS fields [28], [29], due to the ability for unsupervised domain adaptation and flexibility.…”

“…Recent years, how to effectively transfer the accurate supervised information of existing data to abundant newly acquired data with similar scenarios is emerging as research hotspot and difficulty in SAR interpretation [27], [28]. There are similar problems in long-term monitoring about BA expansion and changing applications.…”