Spectral Reconstruction Network From Multispectral Images to Hyperspectral Images: A Multitemporal Case

Li, Tianshuai; Liu, Tianzhu; Wang, Yukun; Li, Xian; Gu, Yanfeng

doi:10.1109/tgrs.2022.3195748

Cited by 10 publications

(2 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, a multitemporal spectral reconstruction network (MTSRN) [31] is proposed to reconstruct HS images from multitemporal MS images, which contains a reconstruction network and a temporal features extraction, and a multitemporal fusion network that can independently reconstruct the MS data of a single-phase into HS data and can improve the reconstruction effect by combining neighboring phase information, respectively. Du et al [32] proposed a novel convolution and transformer joint network (CTJN) including cascaded shallow-feature extraction modules (SFEMs) and deep-feature extraction modules (DFEMs), which can explore local spatial features and global spectral features.…”

Section: Spectral Super-resolution Based On Deep Learningmentioning

confidence: 99%

A Multi-Hyperspectral Image Collaborative Mapping Model Based on Adaptive Learning for Fine Classification

Zhang,

Liu,

Zhang

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

Hyperspectral (HS) data, encompassing hundreds of spectral channels for the same area, offer a wealth of spectral information and are increasingly utilized across various fields. However, their limitations in spatial resolution and imaging width pose challenges for precise recognition and fine classification in large scenes. Conversely, multispectral (MS) data excel in providing spatial details for vast landscapes but lack spectral precision. In this article, we proposed an adaptive learning-based mapping model, including an image fusion module, spectral super-resolution network, and adaptive learning network. Spectral super-resolution networks learn the mapping between multispectral and hyperspectral images based on the attention mechanism. The image fusion module leverages spatial and spectral consistency in training data, providing pseudo labels for spectral super-resolution training. And the adaptive learning network incorporates spectral response priors via unsupervised learning, adjusting the output of the super-resolution network to preserve spectral information in reconstructed data. Through the experiment, the model eliminates the need for the manual setting of image prior information and complex parameter selection, and can adjust the network structure and parameters dynamically, eventually enhancing the reconstructed image quality, and enabling the fine classification of large-scale scenes with high spatial resolution. Compared with the recent dictionary learning and deep learning spectral super-resolution methods, our approach exhibits superior performance in terms of both image similarity and classification accuracy.

show abstract

Section: Spectral Super-resolution Based On Deep Learningmentioning

confidence: 99%

A Multi-Hyperspectral Image Collaborative Mapping Model Based on Adaptive Learning for Fine Classification

Zhang,

Liu,

Zhang

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…However, large bandwidth remote sensing satellite data transmission will inevitably lead to the extremely high data sampling rate at the receiver [6], [7]. In this high-speed environment, even a simple filtering operation will lead to the saturation of the hardware processing module, which cannot meet the requirements of real-time demodulation [8], [9]. Therefore, the high parallel of the demodulation module of remote sensing satellite data ground receiver has become an inevitable development trend.…”

mentioning

confidence: 99%

Optimization of Multi-Subband Parallel and Signal Reconstruction for Remote Sensing Satellite Data Transmission

Wang,

He,

Yang

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

The remote sensing satellite is developing toward high resolution, large data capacity, and fast transmission rate. The ground receiver is correspondingly required to have the high parallel to improve real-time reception and processing capabilities. In the frequency domain, the parallel processing of multiple subbands is achieved by dividing broadband into narrow bands. However, narrow-band subbands are reconstructed into broadband signals, which can lead to cross-subband distortion. It will affect the accuracy of high-resolution remote sensing images. In this article, the subband division and reconstruction framework is proposed by combining the analog filter with the digital filter. The phase calibration methods and digital filter optimization are proposed to improve the amplitude and phase consistency of the reconstructed signal. The simulation results show that the proposed amplitude consistency optimization method effectively reduces the reconstruction error within 0.001 dB. The proposed phase calibration method effectively reduces the bit error rate of the reconstructed signal. The maximum deviation is no more than 0.1%. Experiments have shown that the optimization method can reduce the distortion error of high-resolution remote sensing images.

show abstract

Spectral-Learning-Based Transformer Network for the Spectral Super-Resolution of Remote-Sensing Degraded Images

Liu

et al. 2023

IEEE Geosci. Remote Sensing Lett.

View full text Add to dashboard Cite

Spectral Reconstruction Network From Multispectral Images to Hyperspectral Images: A Multitemporal Case

Cited by 10 publications

References 65 publications

A Multi-Hyperspectral Image Collaborative Mapping Model Based on Adaptive Learning for Fine Classification

A Multi-Hyperspectral Image Collaborative Mapping Model Based on Adaptive Learning for Fine Classification

Optimization of Multi-Subband Parallel and Signal Reconstruction for Remote Sensing Satellite Data Transmission

Spectral-Learning-Based Transformer Network for the Spectral Super-Resolution of Remote-Sensing Degraded Images

Contact Info

Product

Resources

About