Unsupervised Change Detection Using Convolutional-Autoencoder Multiresolution Features

• Published in: IEEE transactions on geoscience and remote sensing Vol. 60; pp. 1 - 19
• Main Authors: Bergamasco, Luca, Saha, Sudipan, Bovolo, Francesca, Bruzzone, Lorenzo
• Format: Journal Article
• Language: English
• Published: New York IEEE 2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Change detection; Convolutional autoencoder (CAE); Data models; Decoding; Deep learning; deep learning (DL); Deforestation; Detection; Feature extraction; Feature maps; Image acquisition; Landsat; Landsat satellites; Methods; multitemporal analysis; Remote sensing; remote sensing (RS); Satellite imagery; Semantics; Task analysis; Training; Transfer learning; unsupervised change detection (CD); unsupervised learning

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 le...