Stella X. Yu : Papers / Google Scholar

Learning Data-Driven Complex-valued Butterfly Transform for Hyperspectral Image Classification Utkarsh Singhal and Stella X. Yu International Joint Conference on Neural Networks (IEEE WCCI), Centro Congressi, Padova, Italy, 18-23 July 2022 Paper | Slides Abstract Hyper-spectral imaging (HSI) is a critical remote sensing modality that captures high-resolution spectral information in addition to high spatial resolution. Due to its ability to capture rich information about the material properties of the target, HSI has found applications such as agriculture, ecological monitoring, urban planning, and medicine. However, HSI images tend to have high spatial resolution and many channels, thus requiring models that can integrate information over a large context and identify spectral signatures. HSI classification datasets are also highly imbalanced, sparsely labeled, and significantly smaller than standard vision datasets like ImageNet or CIFAR, thus motivating smaller models with high sample efficiency. This work combines the strengths of multi-scale representations and data-driven feature learning. We generalize the butterfly transform to a learned complex-valued butterfly layer, allowing for parameter-efficient extraction of hierarchical complex-valued features from 1D signals. This allows us to create a lean yet highly accurate hyperspectral image classification model. Benchmarked on the Indian Pines, ROSIS-03 Pavia University, and Salinas datasets, our method demonstrates accuracy on par with SSDGL while using \textbf{7x fewer} parameters. On the most imbalanced and sparsely labeled dataset, our method outperforms SSDGL. Keywords complex-valued deep learning, FFT, butterfly transform