Self-Guided Autoencoders for Unsupervised Change Detection in Heterogeneous Remote Sensing Images

• Published in: IEEE transactions on artificial intelligence 2024-06, Vol.5 (6), p.2458-2471
• Main Authors: Shi, Jiao, Wu, Tiancheng, Kai Qin, Alex, Lei, Yu, Jeon, Gwanggil
• Format: Article
• Language: English
• Subjects: Autoencoder network; Collaboration; Feature extraction; heterogeneous images; Interference; Ions; Remote sensing; self-guided learning; Task analysis; Training; unsupervised change detection (CD)
• ISSN: 2691-4581; 2691-4581
• DOI: 10.1109/TAI.2024.3357667

To address the problem of enormous differences in two heterogeneous images, the traditional unsupervised frameworks are most normally realized by converting two images into a common domain with various auxiliary strategies, such as transformation and alignment, which requires extensive calculation and has difficulty balancing the training tasks. To achieve a concise framework, this article proposes self-guided autoencoders (SGAE) for unsupervised change detection (CD) in heterogeneous remote sensing images (RSIs). Unlike traditional methods that aim to narrow the differences of heterogeneous images to highlight the changed information, SGAE forces the flow of identification in formation generated in unlabeled data through self-guided iterations. First, initial unsupervised networks output an elementary change map that will be screened to obtain reliable pseudolabels. The selected pseudolabeled samples will be used as the input of a supervised network to obtain another change map. Then, multiple change maps will be fused to refine the confidence of pseudolabels again, obtaining new fused pseudolabeled samples for the self-guided network, which will be trained with pseudolabeled samples and unlabeled samples. Finally, the above operations will be repeated to continuously optimize the net, which helps itself to extract the discriminative features for classification in self-guided iterations. Experiments compared with several algorithms on four datasets demonstrate the effectiveness and robustness of our method, which can help unsupervised models improve discriminative feature extraction and classification performance with a more flexible learning method.