PhyspeNet: An empirical physics-aware network for adaptive speckle reconstructive spectrometry
PhyspeNet：一个基于实证的物理感知网络，用于自适应散斑重建光谱学
PhyspeNet：適応的スペックル再構成分光法を実現する経験的物理意識ネットワーク
PhyspeNet: 적응형 스펙클 재구성 분광법을 위한 경험적 물리 인식 네트워크
PhyspeNet: una red empírica consciente de la física para espectrometría reconstructiva de moteado adaptativa
PhyspeNet : Un réseau empirique sensible à la physique pour la spectrométrie adaptative de reconstruction de taches
PhyspeNet: эмпирическая физически-ориентированная сеть для адаптивной спекл-реконструктивной спектроскопии
Junrui Liang ¹, Min Jiang ², Jun Li ¹, Zhongming Huang ¹, Junhong He ¹, Yanting Guo ¹, Yanzhao Ke ¹, Jun Ye ¹ ³ ⁴, Jiangming Xu ¹, Jinyong Leng ¹ ³ ⁴, Pu Zhou ¹
¹ College of Advanced Interdisciplinary Studies, National University of Defense Technology, Changsha 410073, China
中国 长沙 国防科技大学前沿交叉学科学院
² Test Center, National University of Defense Technology, Xi'an 710106, China
中国 西安 国防科技大学试验训练基地
³ Nanhu Laser Laboratory, National University of Defense Technology, Changsha 410073, China
中国 长沙 国防科技大学南湖之光实验室
⁴ Hunan Provincial Key Laboratory of High Energy Laser Technology, National University of Defense Technology, Changsha 410073, China
中国 长沙 国防科技大学高能激光技术湖南省重点实验室
The speckle reconstructive spectrometer (RS) is revolutionizing the design paradigm of spectrometers, shifting from hardware-dominated architectures to algorithmically driven, computing-focused methodologies. However, traditional spectral reconstruction algorithms with pre-defined regularizes suffer from suboptimal adaptability, and similarly, deep learning methods struggle to handle unseen data types once deployed.
Here we propose a physics-aware spectral reconstruction framework named PhyspeNet, which consists of a convolutional neural network (CNN) and an empirical physics model, facilitating adaptive reconstruction of multiple spectral types without pre-training. The inherent structural priority in the CNN architecture provides an adaptive and remarkably powerful regularization, while the embedded empirical model endows our framework with generality across various dispersive devices, even those exhibiting complex light propagation beyond the scope of analytical models.
By leveraging PhyspeNet, we attain a resolving power of 7×10⁵, surpassing existing spectral reconstruction neural networks trained on polychromatic data. A 700 nm operating range in the near-infrared region is also realized, which, to the best of our knowledge, stands as the widest operational bandwidth ever reported in speckle spectrometry. We believe that this work constitutes a solid step toward adaptive speckle RSs and paves the way for advances in diverse fields, including high-dimensional light field detection, biomedicine and so on.
