Hypervision: An on-chip hyperspectral microsystem for online video-rate computational imaging
- 27 August 2026
- journal article
- research article
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 393 (6814) , 888-894
Abstract
In this work, we tackled the long-standing challenge of the massive computation for hyperspectral imaging that is required to reconstruct and process large-volume spatial-spectral data cubes. Specifically, we designed a hardware accelerator, fabricated as a neural processing unit (NPU) capable of 9.3 tera operations per second at 16-bit integer (INT16), alongside a topology-aware structured pruning strategy for a lightweight reconstruction network. Through integration with our HyperspecI sensor, we demonstrate a fully standalone visible–near-infrared hyperspectral microsystem (~950 grams) that requires neither external power nor computing resources. The microsystem achieved real-time hyperspectral imaging at 32.9 frames per second (512×512, 61 channels) or 24.6 frames per second (1024×1024, 16 channels) and consumed only ~25.3 watts (367 giga–operations per second per watt). Application demonstrations in intelligent driving and air-to-ground monitoring highlight its practical potential advancing computational hyperspectral imaging from offline processing to integrated online perception.Keywords
Funding Information
- National Natural Science Foundation of China (61827901, 62322502)
- Beijing Natural Science Foundation (JQ25017)
- Scientific Research Innovation Capability Support Project for Young Faculty (ZYGXONJSKYCXNLZCXM-I4)
This publication has 82 references indexed in Scilit:
- A systematic review of hyperspectral imaging in precision agriculture: Analysis of its current state and future prospectsComputers and Electronics in Agriculture, 2024
- Meta‐Attention Network Based Spectral Reconstruction with Snapshot Near‐Infrared MetasurfaceAdvanced Materials, 2024
- Handheld snapshot multi-spectral camera at tens-of-megapixel resolutionNature Communications, 2023
- DiVIT: Algorithm and architecture co-design of differential attention in vision transformerJournal of Systems Architecture, 2022
- Deeply learned broadband encoding stochastic hyperspectral imagingLight: Science & Applications, 2021
- End-to-End Low Cost Compressive Spectral Imaging with Spatial-Spectral Self-Attention2020
- A clinically translatable hyperspectral endoscopy (HySE) system for imaging the gastrointestinal tractNature Communications, 2019
- Single-shot on-chip spectral sensors based on photonic crystal slabsNature Communications, 2019
- Sparse Recovery of Hyperspectral Signal from Natural RGB Images2016
- Training-Based Spectral Reconstruction from a Single RGB Image2014