Ensemble Alignment Subspace Adaptation Method for Cross-Scene Classification

An ensemble alignment subspace adaptation (EASA) method is proposed in this letter for the cross-scene classification. It can settle the problem of both foreign objects in the same spectrum and different spectra. The algorithm combines the idea of ensemble learning with the domain adaptive (DA) algorithm. Considering the sample imbalance problem of the original data (OD), the source data (SD) are obtained by multiple random sampling of OD according to certain rules and used as input. Then, geometric alignment and statistical alignment of SD and target data (TD) are performed to build a communal subspace, followed by the classification of TD. The classification labels are finally ensembled by counting the multiple classification results with retaining valid information. This technique can reduce the uncertainty and randomness of generating subspace projections. The experimental results on two real datasets show that the proposed algorithm has a terrific accuracy improvement compared with the traditional machine learning and DA methods.

Automated summary

In this letter, an ensemble alignment subspace adaptation (EASA) method is proposed for cross-scene classification. It addresses the problem of foreign objects in the same spectrum and different spectra by combining ensemble learning with domain adaptive algorithm. The proposed algorithm reduces uncertainty and randomness of subspace projections, and achieves a significant accuracy improvement compared to traditional machine learning and domain adaptation methods as shown in experimental results on two real datasets.