Multi-Spectral Image Classification with Ultra-Lean Complex-Valued Models
This work addresses the challenge of MSI classification for remote sensing applications like disaster recovery, where large-scale annotations are scarce, though it is incremental as it adapts existing complex-valued methods to a new data type.
The paper tackled the problem of classifying multi-spectral imagery (MSI) with limited annotations by applying complex-valued co-domain symmetric models to real-valued MSI data, achieving results that outperform ResNet with data augmentation and modified transfer learning on the xView dataset.
Multi-spectral imagery is invaluable for remote sensing due to different spectral signatures exhibited by materials that often appear identical in greyscale and RGB imagery. Paired with modern deep learning methods, this modality has great potential utility in a variety of remote sensing applications, such as humanitarian assistance and disaster recovery efforts. State-of-the-art deep learning methods have greatly benefited from large-scale annotations like in ImageNet, but existing MSI image datasets lack annotations at a similar scale. As an alternative to transfer learning on such data with few annotations, we apply complex-valued co-domain symmetric models to classify real-valued MSI images. Our experiments on 8-band xView data show that our ultra-lean model trained on xView from scratch without data augmentations can outperform ResNet with data augmentation and modified transfer learning on xView. Our work is the first to demonstrate the value of complex-valued deep learning on real-valued MSI data.