计算机科学
光学滤波器
人工智能
光学
滤波器(信号处理)
傅里叶变换
频道(广播)
卷积神经网络
线性滤波器
光谱包络
光谱分辨率
频域
算法
物理
计算机视觉
谱线
电信
量子力学
天文
作者
Qiwei Li,Jiawei Song,Andrey Alenin,J. Scott Tyo
出处
期刊:Optics Letters
[The Optical Society]
日期:2021-09-01
卷期号:46 (17): 4394-4394
被引量:9
摘要
Channeled spectropolarimetry (CSP) employing low-pass channel extraction filters suffers from cross talk and spectral resolution loss. These are aggravated by empirically defining the shape and scope of the filters for different measured. Here, we propose a convolutional deep-neural-network-based channel filtering framework for spectrally–temporally modulated CSP. The network is trained to adaptively predict spectral magnitude filters (SMFs) that possess wide bandwidths and anti-cross-talk features that adapt to scene data in the two-dimensional Fourier domain. Mixed filters that combine the advantages of low-pass filters and SMFs demonstrate superior performance in reconstruction accuracy.
科研通智能强力驱动
Strongly Powered by AbleSci AI