论文阅读 [TPAMI-2022] Joint Camera Spectral Response Selection and Hyperspectral Image Recovery

论文阅读 [TPAMI-2022] Joint Camera Spectral Response Selection and Hyperspectral Image Recovery

论文搜索(studyai.com)

搜索论文: Joint Camera Spectral Response Selection and Hyperspectral Image Recovery

搜索论文: http://www.studyai.com/search/whole-site/?q=Joint+Camera+Spectral+Response+Selection+and+Hyperspectral+Image+Recovery

关键字(Keywords)

Hyperspectral imaging; Cameras; Training; Spatial resolution; Encoding; Lighting; Pattern analysis; Camera spectral response selection; hyperspectral image recovery; spectral nonlinear mapping; spatial similarity; classification

摘要(Abstract)

Hyperspectral image (HSI) recovery from a single RGB image has attracted much attention, whose performance has recently been shown to be sensitive to the camera spectral response (CSR).

从单个RGB图像中恢复高光谱图像（HSI）引起了广泛关注，其性能最近被证明对相机光谱响应（CSR）非常敏感。.

In this paper, we present an efficient convolutional neural network (CNN) based method, which can jointly select the optimal CSR from a candidate dataset and learn a mapping to recover HSI from a single RGB image captured with this algorithmically selected camera under multi-chip or single-chip setups.

在本文中，我们提出了一种基于卷积神经网络（CNN）的有效方法，该方法可以从候选数据集中联合选择最佳CSR，并学习一个映射，以从该算法选择的相机在多芯片或单芯片设置下捕获的单个RGB图像恢复HSI。.

Given a specific CSR, we first present a HSI recovery network, which accounts for the underlying characteristics of the HSI, including spectral nonlinear mapping and spatial similarity.

鉴于特定的CSR，我们首先提出了HSI恢复网络，该网络解释了HSI的基本特征，包括光谱非线性映射和空间相似性。.

Later, we append a CSR selection layer onto the recovery network, and the optimal CSR under both multi-chip and single-chip setups can thus be automatically determined from the network weights under the nonnegative sparse constraint.

之后，我们在恢复网络上附加一个CSR选择层，从而可以根据非负稀疏约束下的网络权重自动确定多芯片和单芯片设置下的最佳CSR。.

Experimental results on three hyperspectral datasets and two camera spectral response datasets demonstrate that our HSI recovery network outperforms state-of-the-art methods in terms of both quantitative metrics and perceptive quality, and the selection layer always returns a CSR consistent to the best one determined by exhaustive search.

在三个高光谱数据集和两个相机光谱响应数据集上的实验结果表明，我们的HSI恢复网络在定量指标和感知质量方面都优于最先进的方法，并且选择层总是返回与穷举搜索确定的最佳CSR一致的CSR。.

Finally, we show that our method can also perform well in the real capture system, and collect a hyperspectral flower dataset to evaluate the effect from HSI recovery on classification problem…

最后，我们证明了我们的方法在真实的捕获系统中也能表现良好，并收集了一个高光谱花卉数据集来评估HSI恢复对分类问题的影响。。.

作者(Authors)

[‘Ying Fu’, ‘Tao Zhang’, ‘Yinqiang Zheng’, ‘Debing Zhang’, ‘Hua Huang’]