Unsupervised Change Detection Using Convolutional-Autoencoder Multiresolution Features

The use of deep learning (DL) methods for change detection (CD) is currently dominated by supervised models that require a large number of labeled samples. However, these samples are difficult to acquire in the multitemporal case. A possible alternative is leveraging methods that exploit transfer learning for CD by reusing DL models pretrained for other tasks. However, the performance of the transfer-learning-based models decreases as much as the target images differ from the ones used for training the model. To overcome this limit, we propose an unsupervised CD method that exploits multiresolution deep feature maps derived by a convolutional autoencoder (CAE). It automatically learns spatial features from the input during the training phase without requiring any labeled data. The proposed method processes the bitemporal images to obtain and compare multiresolution bitemporal feature maps. These feature maps are then analyzed by a feature-selection technique to select the most discriminant ones. Furthermore, an aggregated multiresolution difference image is computed and used for a detail-preserving multiscale CD. In the context of this CD approach, we propose two alternative strategies to retrieve multiscale reliability maps. We tested the proposed method on bitemporal multispectral images acquired by Landsat-5 and Landsat-8 representing burned areas and Sentinel-2 images representing deforested areas. Results confirm the effectiveness of the proposed CD technique.

Automated summary

The use of supervised deep learning models for change detection requires a large number of labeled samples, which are difficult to obtain in the multitemporal case. To address this issue, a proposed unsupervised method based on convolutional autoencoders can automatically learn spatial features and achieve effective change detection through feature selection and multiscale strategies.