相关实验视频

Updated: Sep 11, 2025

Simultaneously Capturing Real-time Images in Two Emission Channels Using a Dual Camera Emission Splitting System: Applications to Cell Adhesion

Simultaneously Capturing Real-time Images in Two Emission Channels Using a Dual Camera Emission Splitting System: Applications to Cell Adhesion

Published on: September 4, 2013

交替方向与交叉光谱注意力展开,此前用于双摄像头压缩超光谱成像

Yubo Dong, Dahua Gao, Danhua Liu

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society

|August 18, 2025

概括

本研究介绍了ADRNN,这是一种用于双摄像头压缩高光谱成像 (DCCHI) 的新型深度展开网络. 通过解CASSI和PAN子问题,ADRNN有效地重建超光谱图像,实现最先进的性能.

科学领域:

计算成像技术的成像
图像重建图像的重建
用光谱成像进行成像.

背景情况:

编码孔径快照光谱成像 (CASSI) 可以捕捉动态光谱场景,但失去空间信息.
双摄像头压缩高光谱成像 (DCCHI) 使用泛色 (PAN) 摄像头来恢复空间细节,但其结构使基于模型的重建复杂化.
由于DCCHI中被破坏的矩阵属性,现有的方法与数据子问题作斗争.

研究的目的:

从DCCHI数据开发高效准确的超光谱图像重建方法.
克服DCCHI系统中传统的基于模型的方法和深度展开网络 (DUN) 的局限性.
提出一个新的深度展开网络,解决DCCHI成像模型的具体挑战.

主要方法:

提出了一个交替方向DUN (ADRNN),将DCCHI模型解为CASSI和PAN子问题.
在这些子问题中,ADRNN在分析上以代方式解决数据项和联合先前项.
引入了跨光谱变压器 (XST),具有跨光谱注意力和分组查询注意力 (GQA),以利用共同的先验并降低计算成本.

主要成果:

拟议的ADRNN方法在高光谱图像重建中展示了最先进的 (SOTA) 性能.
对模拟和真实DCCHI数据集的实验验证了ADRNN和XST方法的有效性.
建立了一个真正的DCCHI系统,并捕获了大规模的数据集用于进一步研究.

结论:

通过解决固有的成像模型挑战,ADRNN有效地从DCCHI数据中重建超谱图像.
集成XST增强了利用压缩HSI和PAN图像之间的共同先验.
开源代码和数据集有助于未来的压缩超光谱成像研究.

更多相关视频

Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters

Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters

Published on: June 18, 2021

Dual-mode Imaging of Cutaneous Tissue Oxygenation and Vascular Function

Dual-mode Imaging of Cutaneous Tissue Oxygenation and Vascular Function

Published on: December 8, 2010

相关概念视频

IR Spectrum Peak Splitting: Symmetric vs Asymmetric Vibrations

IR Spectrum Peak Splitting: Symmetric vs Asymmetric Vibrations

Identical bonds within a polyatomic group can stretch symmetrically (in-phase) or asymmetrically (out-of-phase). Similar to hydrogen bonding, these vibrations also influence the shape of the IR peak. Generally, asymmetric stretching frequencies are higher than symmetric stretching frequencies. For example, primary amines exhibit two distinct IR peaks between 3300–3500 cm−1 corresponding to the symmetric and asymmetric N-H stretching, while secondary amines exhibit a single...

Upsampling

Upsampling

Managing signal sampling rates is essential in digital signal processing to maintain signal integrity. A decimated signal, characterized by a reduced frequency range due to its lower sampling rate, can be upsampled by inserting zeros between each sample. This upsampling process expands the original spectrum and introduces repeated spectral replicas at intervals dictated by the new Nyquist frequency. To refine this zero-inserted sequence, it is passed through a lowpass filter with a cutoff...