Hyperspectral Resolution Research Articles

Research on Calculation Method of On-Orbit Instrumental Line Shape Function for the Greenhouse Gases Monitoring Instrument on the GaoFen-5B Satellite

The Greenhouse Gases Monitoring Instrument is based on the spectroscopic principle of spatial heterodyne spectroscopy technology and has the characteristics of no moving parts, a hyperspectral resolution, and a large luminous flux. The instrumental line shape function is one of the most important parameters characterizing the features of the instrument, and it plays a vital role in the system error analysis of the instrument’s measurements. To accurately obtain the instrumental line shape function of a spatial heterodyne spectrometer during the on-orbit period and improve the accuracy of gas concentration retrieval, this study develops a method to model and characterize the characteristics of the instrumental line shape function, including modulation loss and phase error. This study employs the solar calibration spectrum in the 1.568–1.583 μm bands to conduct iterative calculations of the instrumental line shape function error model. After the instrumental line function is updated, the average relative deviation is reduced from 1.83% to 0.756% between the theoretical and measured solar spectra. Additionally, the average relative deviation is reduced from 7.049% to 2.106% between the GMI nadir and theoretical nadir spectra. The findings demonstrate that updating the instrumental line shape function mitigates the impact of variations in the spectrometer’s instrumental line shape due to alterations in the orbital environment. This study offers a dependable reference for both the enhancement and oversight of a spectrometer’s instrumental line shape function, along with an investigation of shifts in instrument parameters.

Prototype development and evaluation of a hyperspectral lidar optical receiving system.

As a new type of active Earth observation technology, airborne hyperspectral lidar combines the advantages of traditional lidar 3D information acquisition and passive hyperspectral imaging technology, and it can achieve integrated imaging detection with a high spatial and hyperspectral resolution. Thus, it has become an important future direction of Earth surface remote sensing technology. This article introduces the design and development of an airborne hyperspectral imaging lidar system. The hyperspectral lidar adopts a focal plane splitting method, combined with an array of 168 optical fibers, to couple wide-spectral-range laser echo signals one by one to the corresponding single tube detector, achieving efficient splitting and precise coupling of supercontinuum laser pulse echo signals. This article proposes a fast synchronous calibration method that is suitable for hyperspectral imaging lidar systems. Results show that the spectral range of the hyperspectral lidar system is 400-900 nm, and the spectral resolution of single-fiber detection is greater than 3 nm. Notably, this article focuses on analyzing the abnormal detection channels based on the calibration results. With the test results of adjacent channels combined, the reason for the abnormal spectral bandwidth of channel 17 is analyzed as an example. This research points out the direction for verifying the design parameters of the hyperspectral lidar prototype and lays an important foundation for airborne flight test of the hyperspectral lidar.

The GeoCarb greenhouse gas retrieval algorithm: simulations and sensitivity to sources of uncertainty

Abstract. The Geostationary Carbon Cycle Observatory (GeoCarb) was selected as NASA's second Earth Venture Mission (EVM-2). The scientific objectives of GeoCarb were to advance our knowledge of the carbon cycle, in particular, land–atmosphere fluxes of the greenhouse gases carbon dioxide (CO2) and methane (CH4) and the effects of these fluxes on the Earth's radiation budget. GeoCarb would retrieve column-integrated dry-air mole fractions of CO2 (XCO2), CH4 (XCH4) and CO (XCO), important for understanding tropospheric chemistry), in addition to solar-induced fluorescence (SIF), from hyperspectral resolution measurements in the O2 A-band at 0.76 µm, the weak CO2 band at 1.6 µm, the strong CO2 band at 2.06 µm, and a CH4/CO band at 2.32 µm. Unlike its predecessors (OCO-2/3, GOSAT-1/2, TROPOMI), GeoCarb would be in a geostationary orbit with a sub-satellite point centered over the Americas. This orbital configuration combined with its high-spatial-resolution imaging capabilities would provide an unprecedented view of these quantities on spatial and temporal scales accurate enough to resolve sources and sinks to improve land–atmosphere CO2 and CH4 flux calculations and reduce the uncertainty of these fluxes. This paper will present a description of the GeoCarb instrument and the L2 retrieval algorithms which will be followed by simulation experiments to determine an error budget for each target gas. Several sources of uncertainty will be explored, including that from the instrument calibration parameters for radiometric gain, the instrument line shape (ILS), the polarization, and the geolocation pointing, in addition to forward model parameters including meteorology and spectroscopy, although there are some other instrument-related sources of uncertainty that are left out for this study, including that from “smile”, the keystone effect, stray light, detector persistence, and scene inhomogeneity. The results indicate that the errors (1σ) are less than the instrument's multi-sounding precision requirements of 1.2 ppm, 10 ppb, and 12 ppb (10 %), for XCO2, XCH4, and XCO, respectively. In particular, when considering the sources of uncertainty separately and in combination (all sources included), we find overall RMSEs of 1.06 ppm for XCO2, 8.2 ppb for XCH4, and 2.5 ppb for XCO, respectively. Additionally, we find that, as expected, errors in XCO2 and XCH4 are dominated by forward model and other systematic errors, while errors in XCO are dominated by measurement noise. It is important to note that the GeoCarb mission was canceled by NASA; however, the instrument is still in development and will be delivered to NASA, in full, with the hope that it will eventually be adopted in a future mission proposal.

Design of UV-band hyperspectral resolution imaging spectrometer with small <i>F</i>-number and high variable ratio

Design of UV-band hyperspectral resolution imaging spectrometer with small <i>F</i>-number and high variable ratio

Predictive Prospecting Using Remote Sensing in a Mountainous Terrestrial Volcanic Area, in Western Bangongco–Nujiang Mineralization Belt, Tibet

The Bangongco–Nujiang metallogenic belt of Tibet is a main suture zone in the Qinghai–Tibet Plateau, which is known as an important porphyry–epithermal–skarn Cu-polymetallic mineralization zone in China. The western part of the Bangongco–Nujiang metallogenic belt exposes several medium high-silica terrestrial alkaline volcanic rocks with strong alteration influenced by collision orogeny. Some research has shown that clues to mineralization such as malachite and gossan are found on the surface. However, volcanic rock areas with varied topography place a huge burden on geological investigation, and the existing research on predicting mineralization is relatively scarce. This paper describes the extraction of alteration mineral information based on medium spatial resolution and hyperspectral resolution images, establishing a spectral library of alteration minerals in this area. By analyzing radar data, digital elevation, and synthesis results of different spectral bands, we combine remote sensing with geographic information technology to establish crater markers. The extraction results from multisource and chemical exploration data are superimposed onto the analysis of mineralization characteristics and geological conditions so as to establish the mineralization signatures for terrestrial volcanic rock areas. Eighteen mineralization prospect areas were identified, which can provide technical support for future mineralization research in this belt.

Application of UAV Push-Broom Hyperspectral Images in Water Quality Assessments for Inland Water Protection: A Case Study of Zhang Wei Xin River in Dezhou Distinct, China

A water quality parameter retrieval scheme based on the UAV push-broom hyperspectral images was designed and validated for assessing the ecological health of Zhang Wei Xin River in Dezhou distinct, China. First, a UAV carrying a push-broom hyperspectral imager that is lightweight and has a small size was used to acquire high spatial and hyperspectral resolution images. Then, the mosaicked reflectance data of the whole river were produced by a seamless image mosaicking method with high geometrical accuracy and spectral fidelity. Next, the in-field measurements of different parameters and the corresponding spectral reflectance from the mosaicked images at the sampling points were used to build the water quality parameter retrieval models for total phosphorus (TP), chlorophyll a (Chla), and total suspended solids (TSS). To validate the model, the retrieval results of the testing sampling points were compared with the measured parameters. The coefficients of determination R2 of TP, Chla, and TSS were 0.886, 0.918, and 0.968, respectively. The retrieved TP, Chla, and TSS maps showed that the water pollution of Zhang Wei Xin River is serious, the total phosphorus exceeds the standard, and the water body is in a state of eutrophication. The UAV-based hyperspectral remote sensing technique provides a cost-effective method for inland water monitoring at a local scale with high accuracy.

Development and Core Technologies for Intelligent SWaP3 Infrared Cameras: A Comprehensive Review and Analysis.

With the development of infrared detection and imaging technology, infrared cameras (IRCs) play an important role in many fields, such as military, industry, and civilian. Additionally, the requirements for the size, performance, and intelligence of IRCs are becoming more and more strict. Consequently, the associated research and development (R&D) of IRCs is gradually focused on the aspects of miniaturization, high performance, intelligence, low power consumption, and low cost, involving many frontier fields, including artificial intelligence, new materials, new optical systems, and electronics systems. In fact, there are continual studies on intelligent SWaP3 IRCs, but unfortunately, a systematic arrangement and analysis are lacking. Therefore, a systematical and comprehensive review for the developments and core technologies of the intelligent SWaP3 IRCs is really needed. In this paper, in terms of the aforementioned requirements, we conduct a review and analysis of current intelligent SWaP3 IRCs based on 90 literature and statistics in recent decades to provide the relevant developers with a helpful reference for facilitating the indicator optimization of intelligent SWaP3 IRCs with new developed technologies. We analyze the development of SWaP3 IRCs in the aspects of lightweight, miniaturization, low price, and high performance, including hyperspectral resolution, high spatial resolution, large field of view (FOV), and wide dynamic elaborately. Moreover, the development in low power consumption and intelligence is also discussed in detail. Additionally, we briefly summarize the primary applications of intelligent SWaP3 IRCs in military, scientific, and civil. Then, the core technologies comprising high-integration, lightweight, hyperspectral imaging (HSI), low-power consumption, as well as the realization of high performance such as high-resolution, high-frame, and wide-dynamic range of SWaP3 IRCs are discussed and analyzed in detail. Finally, we prospect for the intelligent SWaP3 IRCs that it is necessary to continuously expand the concept of SWaP3 by reliability, stability, extensibility, and safety. In addition, it is useful to embed cutting-edge technologies such as small pixel pitch array, multi-sensors fusion, and deploy intelligent algorithms to IRCs. Additionally, the improvement of the whole machine from multi-dimension such as chip, camera, and system is expected and needs to be taken more seriously. It is hoped that this paper can provide a reference for the R&D of intelligent SWaP3 IRCs in the future.

Grayscale-patterned integrated multilayer-metal-dielectric microcavities for on-chip multi/hyperspectral imaging in the extended visible bandwidth.

Pixelated filter arrays of Fabry-Perot (FP) cavities are widely integrated with photodetectors to achieve a WYSIWYG ("what you see is what you get") on-chip spectral measurements. However, FP-filter-based spectral sensors typically have a trade-off between their spectral resolution and working bandwidth due to design limitations of conventional metal or dielectric multilayer microcavities. Here, we propose a new idea of integrated color filter arrays (CFAs) consisting of multilayer metal-dielectric-mirror FP microcavities that, enable a hyperspectral resolution over an extended visible bandwidth (∼300 nm). By introducing another two dielectric layers on the metallic film, the broadband reflectance of the FP-cavity mirror was greatly enhanced, accompanied by as-flat-as-possible reflection-phase dispersion. This resulted in balanced spectral resolution (∼10 nm) and spectral bandwidth from 450 nm to 750 nm. In the experiment, we used a one-step rapid manufacturing process by using grayscale e-beam lithography. A 16-channel (4 × 4) CFA was fabricated and demonstrated on-chip spectral imaging with a CMOS sensor and an impressive identification capability. Our results provide an attractive method for developing high-performance spectral sensors and have potential commercial applications by extending the utility of low-cost manufacturing process.

Hyperspectral Image Super-resolution via Knowledge-Driven Deep Unrolling and Transformer Embedded Convolutional Recurrent Neural Network.

Hyperspectral (HS) imaging has been widely used in various real application problems. However, due to the hardware limitations, the obtained HS images usually have low spatial resolution, which could obviously degrade their performance. Through fusing a low spatial resolution HS image with a high spatial resolution auxiliary image (e.g., multispectral, RGB or panchromatic image), the so-called HS image fusion has underpinned much of recent progress in enhancing the spatial resolution of HS image. Nonetheless, a corresponding well registered auxiliary image cannot always be available in some real situations. To remedy this issue, we propose in this paper a newly single HS image super-resolution method based on a novel knowledge-driven deep unrolling technique. Precisely, we first propose a maximum a posterior based energy model with implicit priors, which can be solved by alternating optimization to determine an elementary iteration mechanism. We then unroll such iteration mechanism with an ingenious Transformer embedded convolutional recurrent neural network in which two structural designs are integrated. That is, the vision Transformer and 3D convolution learn the implicit spatial-spectral priors, and the recurrent hidden connections over iterations model the recurrence of the iterative reconstruction stages. Thus, an effective knowledge-driven, end-to-end and data-dependent HS image super-resolution framework can be successfully attained. Extensive experiments on three HS image datasets demonstrate the superiority of the proposed method over several state-of-the-art HS image super-resolution methods.

Dynamic Hyperspectral Pansharpening CNNs

Hyperspectral (HS) pansharpening seeks to integrate low spatial resolution HS (LRHS) images with connected panchromatic (PAN) images to produce high spatial resolution HS (HRHS) images. Traditional pansharpening convolutional neural networks (CNNs) directly map LRHS and PAN images into HRHS images under fixed network parameters, which imply static pansharpening rules. However, real-world HS data are often characterized by spatial variations and intuitively the pansharpening rules should be dynamic. To deal with the dilemma, in this paper we develop dynamic HS pansharpening CNNs. We first specify the concepts of dynamic pansharpening and static pansharpening. Then, we propose a learn-to-learn oriented pansharpening CNN paradigm, which aims to learn a how-to-learn rule to produce spatially adaptive pansharpening rules and comprises three stages of preliminary fusion, scene-sensitive modulation and spectral reconstruction. Finally, following the paradigm, we design two groups of dynamic pansharpening CNNs (DyPNNs), i.e., internal-connection-based and external-connection-based. They involve various spatial modulations, including spatial affine transform (AT), spatial dynamic convolution (DC) or improved spatial attention (SA), and thus consist of six specific DyPNNs: IC-AT-DyPNN, IC-DC-DyPNN, IC-SA-DyPNN, EC-AT-DyPNN, EC-DC-DyPNN and EC-SA-DyPNN. Experimental results on several HS datasets verify the effectiveness of the proposed DyPNNs in terms of both the spatial reconstruction and spectral fidelity.

Research on Compressed Sensing Spectrometry Based on Electro-Optical Transmittance Coding

To develop fast and integrated spectrometry, we present a new method of compressed sensing spectrometry based on electro-optical transmittance coding. A LiNbO3 electro-optic modulator was applied to a directly compressed measurement code in the spectral dimension. Subsequently, a sequential forward floating selection algorithm was employed to select the coding measurement matrix, and sparsity adaptive matching pursuit was used as a solver algorithm. The principle was analyzed, and the measurement system was built for verification experiments. The experimental results reveal that the spectral reconstruction relative error is in the order of 10−2, and the full width at half maximum of the spectral measurement is as high as 1.2 nm. The spectral resolution can reach approximately 0.4 nm in the visible light band range (0.38–0.78 μm) with 1024 spectral channels. The compressed ratio of the compressed sensing spectrometry is up to 1:29. Furthermore, the spectral signal measurement time is only 0.25 ms. The study demonstrates a novel method, which exhibits high precision, speed, high compressed ratio and hyperspectral resolution.

A Dual-Band Model for the Vertical Distribution of Photosynthetically Available Radiation (PAR) in Stratified Waters

Based on the optical properties of water constituents, the vertical variation of photosynthetically available radiation (PAR) can be well modeled with hyperspectral resolution; the intensive computing load, however, demands simplified modeling that can be easily embedded in marine physical and biogeochemical models. While the vertical PAR profile in homogeneous waters can now be accurately modeled with simple parameterization, it is still a big challenge to model the PAR profile in stratified waters with limited variables. In this study, based on empirical equations and simulations, we propose a dual-band model to characterize the vertical distribution of PAR using the chlorophyll concentration (Chl). With an inclusive dataset including cruise data collected in the Southeast Pacific and BGC-Argo data in the global ocean, the model was thoroughly evaluated for its general applicability in three aspects: 1) estimating the entire PAR profile from sea-surface PAR and the Chl profile, 2) estimating the euphotic layer depth from the Chl profile, and 3) estimating PAR just below the sea surface from in situ radiometry measurements. It is demonstrated that the proposed dual-band model is capable of generating similar estimates as that from a hyperspectral model, thus offering an effective module that can be incorporated in large-scale ecosystem and/or circulation models for efficient calculations.

Unsupervised Hyperspectral Microscopic Image Segmentation Using Deep Embedded Clustering Algorithm.

Hyperspectral microscopy in biology and minerals, unsupervised deep learning neural network denoising SRS photos: hyperspectral resolution enhancement and denoising one hyperspectral picture is enough to teach unsupervised method. An intuitive chemical species map for a lithium ore sample is produced using k-means clustering. Many researchers are now interested in biosignals. Uncertainty limits the algorithms' capacity to evaluate these signals for further information. Even while AI systems can answer puzzles, they remain limited. Deep learning is used when machine learning is inefficient. Supervised learning needs a lot of data. Deep learning is vital in modern AI. Supervised learning requires a large labeled dataset. The selection of parameters prevents over- or underfitting. Unsupervised learning is used to overcome the challenges outlined above (performed by the clustering algorithm). To accomplish this, two processing processes were used: (1) utilizing nonlinear deep learning networks to turn data into a latent feature space (Z). The Kullback–Leibler divergence is used to test the objective function convergence. This article explores a novel research on hyperspectral microscopic picture using deep learning and effective unsupervised learning.

Above-Ground Biomass Estimation of Plantation with Different Tree Species Using Airborne LiDAR and Hyperspectral Data

Forest above-ground biomass (AGB) is an important index to evaluate forest carbon sequestration capacity, which is very important to maintain the stability of forest ecosystems. At present, the wide use of remote sensing technology makes it possible to estimate the large-scale forest AGB accurately and efficiently. Airborne hyperspectral remote sensing data can obtain rich spectral information and spatial structure information on the forest canopy with the characteristics of high spatial and hyperspectral resolution. Airborne LiDAR data can describe the three-dimensional structure characteristics of a forest and provide vertical structure information related to biomass. Based on the characteristics of the two data sources, this study takes Gaofeng forest farm in Nanning, Guangxi, as the study area, Chinese fir, pine tree, eucalyptus and other broadleaved trees as the research object, and constructs the AGB estimation models of different tree species by fusing airborne LiDAR and hyperspectral features. Firstly, spectral features, texture features, vegetation index, wavelet transform features and edge features are extracted from hyperspectral data. Canopy structure features, point cloud structure features, point cloud density features and terrain features are extracted from airborne LiDAR data. Secondly, the random forest (RF) method is used to screen the features of the two sets of data, and the features with the highest importance are selected. Finally, based on the optimal features of the two data sources, the forest AGB model is constructed using the multiple stepwise regression method. The results show that the texture features extracted by wavelet transform can be used for AGB modeling. The AGB of eucalyptus has high correlation with height features derived from airborne LiDAR, the AGB of other broadleaved trees mostly depends on the wavelet transform texture features from airborne hyperspectral data, while the AGB of Chinese fir and pine tree has high correlation with both height features and spectral features. Feature-fusion-based LiDAR and hyperspectral data can greatly improve the accuracy of the AGB models. The accuracy of the optimal AGB models of Chinese fir, pine tree, eucalyptus and other broadleaved trees is 0.78, 0.95, 0.72 and 0.89, respectively. In conclusion, more accurate estimation results can be obtained by combining active and passive remote sensing data for forest AGB estimation, which provides a solution for carbon storage assessment and forest ecosystem assessment.

Hyperspectral microscopy imaging based on Fourier ptychographic microscopy

Hyperspectral resolution, high spatial resolution, and a wide field of view (FOV) are the targets of optical spectral microscopy imaging. However, hyperspectral microscopy imaging technology cannot provide a wide FOV and a high spatial resolution at the same time. Fourier ptychographic microscopy (FPM) is a novel microscopy imaging technique that uses LEDs at varying angles to capture a series of low-spatial-resolution images that are used to recover images that have both high spatial resolution and a wide FOV. Since FPM cannot obtain the spectral resolution of the sample, in this paper, an efficient strategy based on the FPM system is proposed for the reconstruction of hyperspectral images. First, the traditional FPM setup is optimized, with a new experimental setup based on halogen lamp illumination and a narrow band-pass filter to capture a series of low-spatial-resolution images at different wavelengths. Second, a new algorithm, combining hyperspectral resolution imaging using interpolation compensation and a phase retrieval algorithm, is proposed to reconstruct high-spatial-resolution, wide FOV, and hyperspectral resolution images. Finally, we verified the feasibility and effectiveness of our experimental setup and algorithm by both simulation and experiment. The results show that our method can not only reconstruct high-spatial-resolution and wide FOV images, but also has a spectral resolution of 5 nm.

Modeling surface ocean phytoplankton pigments from hyperspectral remote sensing reflectance on global scales

Phytoplankton community composition impacts food webs, climate, and fisheries on regional and global scales, and can be assessed at coarse taxonomic resolution from biomarker pigments measured using high-performance liquid chromatography (HPLC). Presently, satellite ocean color provides unprecedented coverage of the global surface ocean and offers reliable estimates of bulk biological properties; however, existing multispectral sensors have limited ability to provide information about phytoplankton community composition. Satellite ocean color at hyperspectral resolution (e.g., NASA's upcoming Plankton, Aerosol, Cloud, and ocean Ecosystem sensor, PACE) is expected to improve estimates of phytoplankton community composition from space. Phytoplankton impact ocean color via contributions to absorption and fluorescence (through phytoplankton pigments) and scattering, especially on narrow spectral scales (5–100 nm). Here, a global open ocean dataset of concurrent HPLC pigments and hyperspectral remote sensing reflectance (Rrs(λ)) observations is used to model phytoplankton pigment composition from optical data. Phytoplankton pigments are reconstructed from Rrs(λ) using optimized principal components regression modeling. This work demonstrates that thirteen phytoplankton pigments, representing five phytoplankton pigment groups (e.g., diatoms, dinoflagellates, haptophytes, green algae, and cyanobacteria), can be modeled from hyperspectral Rrs(λ). Spectral information needed to model each phytoplankton pigment concentration is found throughout the entire visible spectrum and the model results are best at high spectral resolution (≤5 nm). The resulting model recreates observed relationships among pigment concentrations, providing support for the designation of five pigment-based phytoplankton groups for the global open ocean. This work represents a step toward developing robust, global spectral models for phytoplankton pigment composition. However, more high-quality data from a wide range of ecosystems and environments are still needed to achieve this goal.

Imaging Considerations From a Geostationary Orbit Using the Short Wavelength Side of the Mid‐Infrared Water Vapor Absorption Band

AbstractHistorically, the long wavelength side (6–7.5 μm) of the mid‐infrared water vapor absorption band has been used for imaging from the geostationary perspective. This began with the 6.4 μm band on Europe's Meteosat‐1. While geostationary sounders for moisture profiling, including China's hyperspectral resolution infrared sounder and a planned sounder from Europe, are or will be measuring the short wavelength side of the water vapor band, this is not the case for geostationary imagers. Shorter wavelength (5–6 μm) spectral bands for imaging applications should be considered for observing moisture in the mid and lower troposphere because of several potential advantages offered by this spectral range. The short wavelength side of the water vapor band contains fewer additional absorbing gases that overlap the water vapor absorption lines. In addition, the shorter wavelengths would show less diffraction blurring which could enable finer spatial resolution for turbulence detection. This study considers some of the differences and potential advantages of the spectral information in the short wavelength side of the water vapor absorption band from the perspective of geostationary imaging. For dry conditions land heating could impact the qualitative use of observed brightness temperatures at 5.1 μm, while the solar reflection component over most clear‐sky scenes is small for a 5.1 μm band and essentially zero for a 5.6 μm band.

Variable Subpixel Convolution Based Arbitrary-Resolution Hyperspectral Pansharpening

Standard hyperspectral (HS) pansharpening rely on fusion to enhance low-resolution HS (LRHS) images to the resolution of their matching panchromatic (PAN) images, whose practical implementation is normally under a stipulation of scale invariance of model across the training phase and the pansharpening phase. By contrast, arbitrary resolution HS (ARHS) pansharpening seeks to pansharpen LRHS images to any user-customized resolutions. For such a new HS pansharpening task, it is not feasible to train and store CNN models for all possible candidate scales, which implies the single model acquired from the training phase should be capable of being generalized to yield HS images with any resolutions in the pansharpening phase. To address the challenge, a novel variable sub-pixel convolution (VSPC)-based CNN (VSPC-CNN) method following our arbitrary upsampling CNN (AU-CNN) framework is developed for ARHS pansharpening. The VSPC-CNN method comprises a two-stage elevating thread. The first stage is to improve the spatial resolution of input HS image to that of the PAN image through a pre-pansharpening module and then a VSPC-encapsulated arbitrary scale attention upsampling (ASAU) module is cascaded for arbitrary resolution adjustment. After training with given scales, it can be generalized to pansharpen HS image to arbitrary scales under the spatial patterns invariance across the training and pansharpening phases. Experimental results from several specific VSPC-CNNs on both simulated and real HS datasets show the superiority of the proposed method.

CNN-Based Hyperspectral Pansharpening With Arbitrary Resolution

Traditional hyperspectral (HS) pansharpening aims at fusing a HS image with its panchromatic (PAN) counterpart, to bring the spatial resolution of the HS image to that of the PAN image. However, in many practical applications, arbitrary resolution HS (ARHS) pansharpening is required, where the HS and PAN images need to be integrated to generate a pansharpened HS image with arbitrary resolution (usually higher than that of the PAN image). Such an innovative task brings forth new challenges for the pansharpening technique, mainly including how to reconstruct HS images beyond the training scale and how to guarantee spectral fidelity at any spatial resolutions. To tackle the challenges, we present a novel convolutional neural network (CNN)-based method for ARHS pansharpening called ARHS-CNN. It is based on a two-step relay optimization process, which is associated with a multilevel enhancement subnetwork and a rescaling subnetwork. With a careful design following the thread, our ARHS-CNN is able to pansharpen HS images to any spatial resolutions using just a single CNN model trained on a limited number of scales while meantime to keep spectral fidelity at those resolutions, which wins an obvious advantage over traditional pansharpening methods. Experimental results obtained on several datasets verify the excellent performance of our ARHS-CNN method.

Digital micromirror device and spatial heterodyne spectroscopy combined modulation interference spectroscopy

Spatial heterodyne spectroscopy (SHS) has advantages of hyperspectral, high luminous flux, and high stability. However, its main performance indexes, such as spectral resolution, bandwidth, dynamic range and signal-to-noise ratio (SNR) are all limited by the performance parameters of the area array detector. In this paper, a digital micromirror device (DMD) is applied to the interference spectroscopy, and DMD and SHS combined modulation interference spectroscopy (DMD–SHS) is proposed. On the basis of retaining advantages of SHS such as high stability and hyperspectral resolution, DMD–SHS uses the linear array detector, eliminating high-performance area-array detector dependence and dramatically reducing the cost of the instrument. It can also improve the instrument’s SNR and dynamic range, and enhance weak signals’ detection capability. DMD–SHS theory is analyzed theoretically and further verified by simulation in this paper. The combined modulation method proposed in the paper can also be applied to other interference spectroscopy, and the related research has important scientific significance and application prospect.