JigsawHSI: a network for Hyperspectral Image classification
Jaime Moraga
TL;DR
JigsawHSI introduces an inception-style hyperspectral image classifier that pairs spectral-focused and spatial-aware pathways to achieve competitive performance against state-of-the-art 3D-CNN approaches. By combining configurable spectral-network-in-network components, center-pixel crop pathways, and dimensionality reduction as a pre-processing step, the method matches or surpasses HybridSN on Indian Pines, Pavia University, and Salinas datasets while using primarily 2D convolutions. The work demonstrates robust per-pixel classification with high overall accuracy and provides a generalized, extensible architecture suitable for multi-modal geoscience data. Public code and a flexible configuration framework further enhance its practical impact for researchers and practitioners in hyperspectral analysis and related domains.
Abstract
This article describes Jigsaw, a convolutional neural network (CNN) used in geosciences and based on Inception but tailored for geoscientific analyses. Introduces JigsawHSI (based on Jigsaw) and uses it on the land-use land-cover (LULC) classification problem with the Indian Pines, Pavia University and Salinas hyperspectral image data sets. The network is compared against HybridSN, a spectral-spatial 3D-CNN followed by 2D-CNN that achieves state-of-the-art results on the datasets. This short article proves that JigsawHSI is able to meet or exceed HybridSN's performance in all three cases. It also introduces a generalized Jigsaw architecture in d-dimensional space for any number of multimodal inputs. Additionally, the use of jigsaw in geosciences is highlighted, while the code and toolkit are made available.