Advanced Hyperspectral Imaging Research Articles

High SNR eSWIR Image Sensor Applied in Advanced Hyperspectral Imager Aboard China’s GaoFen-5 Satellite and ZY-1 Satellite

This paper presents an extended short-wave infrared (eSWIR) image sensor, a large format infrared focal plane array (IRFPA), applied in advanced hyperspectral imager (AHSI) aboard China’s three GaoFen-5 satellites and two more followed ZY-1 satellites. Hyperspectral imaging has been emerged as a very important application in Earth observation instruments, for the spectral detection provides the spectrum information of ground objects except for the spatial imaging of the objects. AHSI has a 60-km swath width and a 30-m spatial resolution, and has high signal-to-noise ratio (SNR) in spectrum of interest. Large format infrared focal plane arrays which are ordinarily used in staring infrared imaging have some special performance requirements when applied in hyperspectral imaging, such as higher quantum efficiency, lower dark current, lower noise and selectable current conversion gain to match the different radiation flux of different subdivided spectral bands. We developed a format of 2048×512 HgCdTe/Si IRFPA sensor (equivalent 2012×256 after pixel overlapping and binning) in which 2012 pixels meet the demands of 60km swath width and 256 pixels line meet the 200 spectral bands. The sensor has 8-level gains selectable by spectral bands, and array dark current densities on an order of 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-10</sup> A/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> , quantum efficiency exceeding 80%, and the operability of 99.5% at operating temperature of around 110K. The SNR of this FPA achieved 150 when illuminated under 5×10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sup> photons/pixel.

A local enhanced mamba network for hyperspectral image classification

Deep learning has significantly advanced hyperspectral image (HSI) classification, primarily due to its robust nonlinear feature extraction capabilities. The vision transformer has achieved notable performance but is limited by the quadratic computational burden of its self-attention mechanism. Recently, a network based on state space model named Mamba, has attracted considerable attention for its linear complexity and commendable performance. Nevertheless, Mamba was originally designed for one-dimensional causal sequence modeling, and its effectiveness in inherent non-causal HSI classification remains to be fully validated. To address this issue, we propose a novel Local Enhanced Mamba (LE-Mamba) network for hyperspectral image classification, which mainly comprises a Local Enhanced Spatial SSM (LES-S6), a Central Region Spectral SSM (CRS-S6), and a Multi-Scale Convolutional Gated Unit (MSCGU). The LES-S6 improves non-causal local feature extraction by incorporating a multi-directional local spatial scanning mechanism. Additionally, the CRS-S6 employs a bidirectional scanning mechanism in the spectral dimension to capture fine spectral details and integrate them with spatial information. The MSCGU utilizes a convolutional gating mechanism to aggregate features from diverse scanning routes and extract high-level semantic information. The overall accuracies of LE-Mamba on Indian Pines, WHU-Hi-HanChuan, WHU-Hi-LongKou, and Pavia University datasets are 99.16 %, 98.16 %, 99.57 %, and 99.63 %, respectively. Extensive experimental results on these four public datasets demonstrate that the LE-Mamba outperforms eight mainstream deep learning models in classification performance.

Predicting Particle Size and Soil Organic Carbon of Soil Profiles Using VIS-NIR-SWIR Hyperspectral Imaging and Machine Learning Models

Modeling spectral reflectance data using machine learning algorithms presents a promising approach for estimating soil attributes. Nevertheless, a comprehensive investigation of the most effective models, parameters, wavelengths, and data acquisition techniques is essential to ensure optimal predictive accuracy. This work aimed to (a) explore the potential of the soil spectral signature obtained in different spectral bands (VIS-NIR, SWIR, and VIS-NIR-SWIR) and, by using hyperspectral imaging and non-imaging sensors, in the predictive modeling of soil attributes; and (b) analyze the accuracy of different ML models in predicting particle size and soil organic carbon (SOC) applied to the spectral signature of different spectral bands. Six soil monoliths, located in the central north region of Parana, Brazil, were collected and scanned via hyperspectral cameras (VIS-NIR camera and SWIR camera) and spectroradiometer (VIS-NIR-SWIR) in the laboratory. The spectral signature of the soils was analyzed and subsequently applied to ML models to predict particle size and SOC. Each set of data obtained by the different sensors was evaluated separately. The algorithms used were k-nearest neighbors (KNN), support vector machine (SVM), random forest (RF), linear regression (LR), artificial neural network (NN), and partial least square regression (PLSR). The most promising predictive performance was observed for the complete VIS-NIR-SWIR spectrum, followed by SWIR and VIS-NIR. Meanwhile, KNN, RF, and NN models were the most promising algorithms in estimating soil attributes for the dataset obtained from both sensors. The general mean R2 (determination coefficient) values obtained using these models, considering the different spectral bands evaluated, were around 0.99, 0.98, and 0.97 for sand prediction, and around 0.99, 0.98, and 0.96 for clay prediction. The lower performances, obtained for the datasets from both sensors, were observed for silt and SOC, with R2 results between 0.40 and 0.59 for these models. KNN demonstrated the best predictive performance. Integrating effective ML models with robust sample databases, obtained by advanced hyperspectral imaging and spectroradiometers, can enhance the accuracy and efficiency of soil attribute prediction.

Advanced hyperspectral image classification via adaptive triplet networks and chaotic quasi oppositional optimization

Advanced hyperspectral image classification via adaptive triplet networks and chaotic quasi oppositional optimization

Graph feature fusion driven by deep autoencoder for advanced hyperspectral image unmixing

In this paper, we propose a pioneering approach for blind Hyperspectral Image (HSI) unmixing named Multi-Features Graph Deep Fusion Learning Networks for HSI Unmixing (MF-GDL). Our method leverages the power of multi-feature graph deep fusion learning networks. MF-GDL integrates spectral HSI graph features with complementary spatial graph features, overcoming the fixed window size limitation observed in prior deep learning approaches in capturing spatial relationships effectively. A key contribution lies in the introduction of multi-view graph learning, a novel aspect in HSI unmixing. By incorporating graph similarity-based spatial information, we address an important gap in the existing literature, marking the first attempt, to our knowledge, to tackle this challenge using a graph deep learning method. Furthermore, we introduce auxiliary information based on Extended Morphological Profiles (EMPs) to enrich spatial context beyond high-level spatial relationships between neighboring pixels. This multi-view fusion learning not only improves accuracy in material abundance estimation but also demonstrates the significance of considering multiple perspectives in HSI unmixing. We validate the efficacy of MF-GDL on four real datasets: Jasper Ridge, Samson, Urban, and Apex showcasing its robust performance and highlighting the substantial contribution of the multi-view aspect in advancing HSI unmixing techniques.

A survey of methane point source emissions from coal mines in Shanxi province of China using AHSI on board Gaofen-5B

Abstract. Satellite-based detection of methane (CH4) point sources is crucial in identifying and mitigating anthropogenic emissions of CH4, a potent greenhouse gas. Previous studies have indicated the presence of CH4 point source emissions from coal mines in Shanxi, China, which is an important source region with large CH4 emissions, but a comprehensive survey has remained elusive. This study aims to conduct a survey of CH4 point sources over Shanxi's coal mines based on observations of the Advanced Hyperspectral Imager (AHSI) on board the Gaofen-5B satellite (GF-5B/AHSI) between 2021 and 2023. The spectral shift in centre wavelength and change in full width at half-maximum (FWHM) from the nominal design values are estimated for all spectral channels, which are used as inputs for retrieving the enhancement of the column-averaged dry-air mole fraction of CH4 (ΔXCH4) using a matched-filter-based algorithm. Our results show that the spectral calibration on GF-5B/AHSI reduced estimation biases of the emission flux rate by up to 5.0 %. We applied the flood-fill algorithm to automatically extract emission plumes from ΔXCH4 maps. We adopted the integrated mass enhancement (IME) model to estimate the emission flux rate values from each CH4 point source. Consequently, we detected CH4 point sources in 32 coal mines with 93 plume events in Shanxi province. The estimated emission flux rate ranges from 761.78 ± 185.00 to 12 729.12 ± 4658.13 kg h−1. Our results show that wind speed is the dominant source of uncertainty contributing about 84.84 % to the total uncertainty in emission flux rate estimation. Interestingly, we found a number of false positive detections due to solar panels that are widely spread in Shanxi. This study also evaluates the accuracy of wind fields in ECMWF ERA5 reanalysis by comparing them with a ground-based meteorological station. We found a large discrepancy, especially in wind direction, suggesting that incorporating local meteorological measurements into the study CH4 point source are important to achieve high accuracy. The study demonstrates that GF-5B/AHSI possesses capabilities for monitoring large CH4 point sources over complex surface characteristics in Shanxi.

Fabrication of a high-quality, small blaze angle grating for visible short-wave infrared hyperspectral cameras

This study outlines the development of a low line density, small blaze angle grating, optimized for a visible to short-wave infrared hyperspectral camera. An analysis of grating specifications was conducted to meet the precise requirements of this application, particularly focusing on the stringent tolerance limits for the blaze angle. A specialized ruling tool adjustment device was designed to adhere to these exacting blaze angle tolerances. The grating groove shape was examined using atomic force microscopy (AFM), and the theoretical diffraction efficiency of the grating was calculated based on these observations. Additionally, laser-based methods were employed to measure the actual diffraction efficiency of the grating, while interferometry was used to assess the grating’s diffraction wavefront. The test results demonstrate our capability to fabricate high-quality gratings with a low line density and small blaze angles that are suitable for advanced hyperspectral imaging applications.

Development of a Novel Implementation of a Remotely Piloted Aircraft System over 25 kg for Hyperspectral Payloads

A main aspect limiting the operation of low-altitude remotely piloted aircraft systems (RPAS) over 25 kg, integrating pushbroom hyperspectral sensors, comes from the challenges related to aircraft performance (e.g., flight time) and regulatory aspects deterring the users from pushing beyond this weight limit. In this study, we showcase a novel implementation using the DJI Agras T30 as an aerial system for integrating an advanced hyperspectral imager (HSI, Hyspex VS-620). We present the design and fabrication approach applied to integrate the HSI payload, the key considerations for powering the HSI and its gimbal, and the results from vibration and wind tunnel tests. We also evaluate the system’s flight capacity and the HSI’s geometric and radiometric data qualities. The final weight of the T30 after the integration of the HSI payload and ancillary hardware was 43 kg. Our vibration test showed that the vibration isolator and the gimbal reduced the vibration transmission to above 15 Hz but also introduced a resonant peak at 9.6 Hz that led to vibration amplification in the low-frequency range near 9.6 Hz (on the order of an RMS of ~0.08 g). The wind tunnel test revealed that the system is stable up to nearly twice the wind speed rating of the manufacturer’s specifications (i.e., 8 m/s). Based on the requirements of the Canadian Special Flight Operations Certificate (RPAS &gt; 25 kg) to land at a minimal battery level of ≥30%, the system was able to cover an area of ~2.25 ha at a speed of 3.7 m/s and an altitude of 100 m above ground level (AGL) in 7 min. The results with the HSI payload at different speeds and altitudes from 50 m to 100 m AGL show hyperspectral imagery with minimal roll–pitch–yaw artefacts prior to geocorrection and consistent spectra when compared to nominal reflectance targets. Finally, we discuss the steps followed to deal with the continuously evolving regulatory framework developed by Transport Canada for systems &gt; 25 kg. Our work advances low-altitude HSI applications and encourages remote sensing scientists to take advantage of national regulatory frameworks, which ultimately improve the overall quality of HSI data and safety of operations with RPAS &gt; 25 kg.

Predicting the Surface Soil Texture of Cultivated Land via Hyperspectral Remote Sensing and Machine Learning: A Case Study in Jianghuai Hilly Area

Soil reflectance spectra and hyperspectral images have great potential to monitor and evaluate soil texture in large-scale scenarios. In hilly areas, sand, clay, and silt have similar spectral characteristics in visible, near-infrared, and short-wave infrared (VNIR-SWIR) reflection spectra. Soil texture spectra belong to mixed spectra despite some differences in particle size, mineral composition, and water content, making their distinction difficult. The accurate identification of the content within different particle sizes is difficult as it involves capturing spectral reflection features. Therefore, this study aimed to predict soil texture content through machine learning and unmixing the soil texture’s spectra while also comparing their respective modelling performances. Taking typical cultivated land in the Jianghuai hills as an example, the GaoFen-5 Advanced Hyperspectral Imaging (GF-5 AHSI) laboratory spectra of soil samples were used to predict sand, silt, and clay particle contents using partial least squares regression (PLSR) and convolutional neural networks (CNNs). The entire spectra of VNIR-SWIR regions were smoothed, and the dimensions were reduced via principal component analysis (PCA). The prediction models of sand, silt, and clay particle content were constructed, and inversion maps were generated using AHSI. The results showed that the PCA-CNN model achieved a higher prediction precision than the PCA-PLSR in both ASD and GF-5 data. Clay content exhibited the highest predictive performance with a coefficient of determination (R2) of 0.948 and 0.908 and a root mean square error (RMSE) of 26.51 g/kg and 31.24 g/kg, respectively, which represented a 39.0% and 79.8% increase in R2 and a 57% and 57.1% decrease in RMSE compared to that of the PCA-PLSR. This method indicates that the PCA-CNN model can effectively achieve nonlinear interactions between multiple spectral components and better model and fit spectral mixing processes; moreover, it provides an alternative method for investigating the spatial distribution of soil texture.

Spatial-Convolution Spectral-Transformer Interactive Network for Large-Scale Fast Refined Land Cover Classification and Mapping Based on ZY1-02D Satellite Hyperspectral Imagery

Satellite hyperspectral imagery is an important data source for large-scale refined land cover classification and mapping, but the high spatial heterogeneity and spectral variability at low spatial resolution and the high computation cost for massive data remain challenges in the research community. In recent years, convolutional neural network (CNN) models with the capability for feature extraction have been widely used in hyperspectral image classification. However, incomplete feature extraction, inappropriate feature fusion, and high time consumption are still the major problems for CNN applications in large-scale fine land cover mapping. In this study, a Spatial-Convolution Spectral-Transformer Interactive Network (SCSTIN) was proposed to integrate 2D-CNN and Transformer into a dual-branch network to enhance feature extraction capabilities by exploring spatial context information and spectral sequence signatures in a targeted manner. In addition, spatial-spectral interactive fusion (SSIF) units and category-adaptive weighting (CAW) as two feature fusion modules were also adopted between and after the two feature extraction branches to improve efficiency in feature fusion. The ZY1-02D hyperspectral imagery was collected to conduct the experiments in the study area of the eastern foothills of the Helan Mountains (EFHLM), covering an area of about 8800 km2, which is the largest hyperspectral dataset as far as we know. To explore the potential of the proposed network in terms of accuracy and efficiency, SCSTIN models with different depths (SCSTIN-4 and SCSTIN-2) were performed. The results suggest that compared with the previous eight advanced hyperspectral image classifiers, both SCSTIN models achieved satisfactory performance in accuracy and efficiency aspects with low complexity, where SCSTIN-4 achieved the highest accuracy and SCSTIN-2 obtained higher efficiency. Accordingly, the SCSTIN models are reliable for large-scale fast refined land cover classification and mapping. In addition, the spatial distribution pattern of diverse ground objects in EFHLM is also analyzed.

Small Sample Hyperspectral Image Classification Based on the Random Patches Network and Recursive Filtering

In recent years, different deep learning frameworks were introduced for hyperspectral image (HSI) classification. However, the proposed network models have a higher model complexity, and do not provide high classification accuracy if few-shot learning is used. This paper presents an HSI classification method that combines random patches network (RPNet) and recursive filtering (RF) to obtain informative deep features. The proposed method first convolves image bands with random patches to extract multi-level deep RPNet features. Thereafter, the RPNet feature set is subjected to dimension reduction through principal component analysis (PCA), and the extracted components are filtered using the RF procedure. Finally, the HSI spectral features and the obtained RPNet-RF features are combined to classify the HSI using a support vector machine (SVM) classifier. In order to test the performance of the proposed RPNet-RF method, some experiments were performed on three widely known datasets using a few training samples for each class, and classification results were compared with those obtained by other advanced HSI classification methods adopted for small training samples. The comparison showed that the RPNet-RF classification is characterized by higher values of such evaluation metrics as overall accuracy and Kappa coefficient.

Soil copper concentration map in mining area generated from AHSI remote sensing imagery

Hyperspectral remote sensing has the advantages to predict and map soil heavy metal concentration over conventional monitoring methods and multispectral remote sensing. In quantitative applications of hyperspectral remote sensing imagery, the contribution of hyperspectral bands is different, and abnormal prediction values resulted from incorrectly classified bare soil images are a major problem. In this study, a variable weighting method was proposed to weight the hyperspectral bands, and a probability threshold was used to improve the classification to mitigate the problem of abnormal prediction values. The variable weighting was conducted by using the absorption depths obtained by continuum removal. Soil samples were collected from a mining area in southwestern China. Hyperspectral remote sensing imagery was acquired by the Advanced Hyperspectral Imager (AHSI) abroad on Geofen-5 (GF-5) satellite. Genetic algorithm and partial least squares regression (PLSR) were adopted to calibrate prediction models. In prediction of soil copper (Cu) concentration, root mean square error (RMSE) and coefficient of determination (R2) were 21.59 mg kg−1 and 0.60 for the prediction using raw reflectance spectra, and the values were improved to 18.33 mg kg−1 and 0.71 by using the weighted reflectance spectra. The developed prediction model was applied to the AHSI imagery to predict Cu concentration in bare soil areas. In prediction of Cu concentration using the AHSI imagery, negative prediction values were eliminated by using the bare soil image extracted by the improved classification. Based on the prediction, soil Cu concentration map was generated by kriging spatial interpolation. The result indicates that the proposed variable weighting method is effective and the problem of abnormal prediction values could be mitigated by using improved bare soil images. Further analysis indicates that some indices with proper thresholds also could be used to get improved bare soil images.

Spectral unmixing analysis for the lunar surface mineralogy using Chandrayaan-2 IIRS imagery

Hyperspectral imagery (HSI) in the visible to near-infrared wavelength region has a high potential for deciphering mineral compositions of terrestrial and planetary surfaces. Thus, ISRO's Imaging Infrared Spectrometer (IIRS) sensor onboard Chandrayaan-2 (Ch-2) orbiter provides an opportunity to utilise the hyperspectral observations to characterise the lunar surface minerals and composition significantly. Hence, a sensitivity study of hyperspectral observations of the IIRS sensor was carried out for the detection and mapping of various lunar minerals using spectral unmixing analysis. A common region within the hyperspectral image obtained from Chandrayaan-1's (Ch-1) Moon Mineralogy Mapper (M3) sensor and Ch-2 IIRS sensor has been identified in the South-East of the Taurus-Littrow valley nearby Gardner crater for comparison and validation purposes. Ch-2 IIRS is an advanced hyperspectral imaging spectrometer capable of collecting spectra over a wider spectral range (800 ​nm ​− ​5000 ​nm) with high spatial (~80 m) and spectral resolution (20 n m ​− ​25 ​n m ). To examine spectral unmixing analysis, an electromagnetic range (800 ​nm ​− ​2500 ​nm) in the near-infrared region was chosen for the Ch-2 IIRS sensor. An L 1 ​− ​ norm based denoising algorithm is used for mixed noise removal from both the IIRS and M3 reflectance datasets. Endmember extraction and fractional abundance estimation are carried out using the N-Findr algorithm and the fully-constrained least square method, respectively. Our results reveal different minerals such as high-calcium pyroxene (Clinopyroxene), low-calcium pyroxene (Orthopyroxene), and regolith surface (matured and younger). The results obtained from the IIRS and M3 observations are in strong agreement. Hence, the hyperspectral observations made by the IIRS sensor are extremely effective for describing the mineral and chemical compositions of the lunar surface. • Spectral unmixing analysis is executed to determine lunar surface mineralogy and their abundance using Hyperspectral data. • For comparative analysis, co-located datasets from Chandrayaan-2 (Ch-2) IIRS and Chandrayaan-1(Ch-1) M3 are investigated. • An L1-norm-based denoising algorithm is introduced to significantly improve the data quality by removing the complex noise. • The methodology can be automated to develop the lunar mineralogical and chemical map in a time-efficient and systematic way.

Hyperspectral Image Super-Resolution Method Based on Spectral Smoothing Prior and Tensor Tubal Row-Sparse Representation

Due to the limited hardware conditions, hyperspectral image (HSI) has a low spatial resolution, while multispectral image (MSI) can gain higher spatial resolution. Therefore, derived from the idea of fusion, we reconstructed HSI with high spatial resolution and spectral resolution from HSI and MSI and put forward an HSI Super-Resolution model based on Spectral Smoothing prior and Tensor tubal row-sparse representation, termed SSTSR. Foremost, nonlocal priors are applied to refine the super-resolution task into reconstructing each nonlocal clustering tensor. Then per nonlocal cluster tensor is decomposed into two sub tensors under the tensor t-prodcut framework, one sub-tensor is called tersor dictionary and the other is called tensor coefficient. Meanwhile, in the process of dictionary learning and sparse coding, spectral smoothing constraint is imposed on the tensor dictionary, and L1,1,2 norm based tubal row-sparse regularizer is enforced on the tensor coefficient to enhance the structured sparsity. With this model, the spatial similarity and spectral similarity of the nonlocal cluster tensor are fully utilized. Finally, the alternating direction method of multipliers (ADMM) was employed to optimize the solution of our method. Experiments on three simulated datasets and one real dataset show that our approach is superior to many advanced HSI super-resolution methods.

Performance of hyperspectral data in predicting and mapping zinc concentration in soil

Reflectance spectroscopy in visible, near-infrared, and short-wave infrared (VNIR-SWIR) region has been recognized as a promising alternative for prediction of heavy metal concentration in soil. Compared with VNIR-SWIR reflectance spectroscopy, VNIR reflectance spectroscopy is less affected by atmospheric water vapor and has relatively high signal to noise ratio. The performances of VNIR and VNIR-SWIR hyperspectral data in predicting and mapping heavy metal concentration in soil were explored. In this study, laboratory spectra of soil samples collected from an agricultural area and Advanced Hyperspectral Imaging (AHSI) remote sensing imagery were used to predict and map zinc (Zn) concentration with genetic algorithm and partial least squares regression (GA-PLSR). The entire spectral regions of VNIR-SWIR and VNIR and spectral subsets extracted from the entire spectral regions were used in the prediction. For the laboratory spectra, the combination of the spectral bands extracted from the absorption features at 500 nm and in 600–800 nm obtained the highest prediction accuracy with the root mean square error (RMSE) and coefficient of determination (R2) values of 8.90 mg kg−1 and 0.72. For soil spectra from AHSI remote sensing imagery, the highest prediction accuracy was achieved by using the spectral bands extracted from the absorption feature in 600–800 nm with the RMSE and R2 values of 9.02 mg kg−1 and 0.75. Soil Zn concentration maps were generated with the established prediction models using AHSI remote sensing imagery. Analysis on the Zn concentration maps shows that the prediction model established using the spectral bands extracted from the absorption feature in 600–800 nm has a better performance in mapping Zn concentration. The results indicate that VNIR hyperspectral data outperforms VNIR-SWIR hyperspectral data in predicting and mapping Zn concentration in soil, which provides an alternative to the application of hyperspectral data in soil science.

Spectral Index for Mapping Topsoil Organic Matter Content Based on ZY1-02D Satellite Hyperspectral Data in Jiangsu Province, China

Estimation of soil organic matter content (SOMC) is essential for soil quality evaluation. Compared with traditional multispectral remote sensing for SOMC mapping, the distribution of SOMC in a certain area can be obtained quickly by using hyperspectral remote sensing data. The Advanced Hyper-Spectral Imager (AHSI) onboard the ZY1-02D satellite can simultaneously obtain spectral information in 166 bands from visible (400 nm) to shortwave infrared (2500 nm), providing an important data source for SOMC mapping. In this study, SOMC-related spectral indices (SIs) suitable for this satellite were analyzed and evaluated in Shuyang County, Jiangsu Province. A series of SIs were constructed for the bare soil and vegetation-covered (mainly rice crops and tree seedlings) areas by combining spectral transformations (such as reciprocal and square root) and dual-band index formulas (such as ratio and difference), respectively. The optimal SIs were determined based on Pearson’s correlation coefficient (ρ) and satellite data quality, and applied to SOMC level mapping and estimation. The results show that: (1) The SI with the highest ρ in the bare soil area is the ratio index of original reflectance at 654 and 679 nm (OR-RI(654,679)), whereas the SI in the vegetation area is the square root of the difference between the reciprocal reflectance at 551 and 1998 nm (V-RR-DSI(551,1998)); (2) the spatial distribution trend of regional SOMC results obtained by linear regression models of OR-RI(654,679) and V-RR-DSI(551,1998) is consistent with the samples; (3) based on the optimal SIs, support vector machine and tree ensembles were used to predict the SOMC of bare soil and vegetation-covered areas of Shuyang County, respectively. The determination coefficient of the soil–vegetation combined prediction results is 0.775, the root mean square error is 3.72 g/kg, and the residual prediction deviation is 2.12. The results show that the proposed SIs for ZY1-02D satellite hyperspectral data are of great potential for SOMC mapping.

Optimization and Evaluation of Widely-Used Total Suspended Matter Concentration Retrieval Methods for ZY1-02D’s AHSI Imagery

Total suspended matter concentration (CTSM) is an important parameter in aquatic ecosystem studies. Compared with multispectral satellite images, the Advanced Hyperspectral Imager (AHSI) carried by the ZY1-02D satellite can capture finer spectral features, and the potential for CTSM retrieval is enormous. In this study, we selected seven typical Chinese inland water bodies as the study areas, and recalibrated and validated 11 empirical models and two semi-analytical models for CTSM retrieval using the AHSI data. The results showed that the semi-analytical algorithm based on the 697 nm AHSI-band achieved the highest retrieval accuracy (R2 = 0.88, average unbiased relative error = 34.43%). This is because the remote sensing reflectance at 697 nm was strongly influenced by CTSM, and the AHSI image spectra were in good agreement with the in-situ spectra. Although further validation is still needed in highly turbid waters, this study shows that AHSI images from the ZY1-02D satellite are well suited for CTSM retrieval in inland waters.

A Spectrum Extension Approach for Radiometric Calibration of the Advanced Hyperspectral Imager Aboard the Gaofen-5 Satellite

The advanced hyperspectral imager (AHSI) is one of the sensors aboard the Chinese Gaofen-5 (GF-5) satellite, possessing characteristics of high spatial and spectral resolution, as well as width swath. To better understand the radiometric performance of GF-5/AHIS after its launch, this article presents an on-orbit radiometric calibration approach for AHSI visible and near-infrared (VNIR) and shortwave infrared (SWIR) sensors from field automatic observations with a field spectrometer in the absence of SWIR measurements. A spectrum extension method was proposed to extend the retrieved surface hyperspectral reflectance in the VNIR spectral ranges to SWIR by incorporating the historical hyperspectral reflectance library. The radiometric calibration coefficients of GF-5/AHSI were calculated by linear fitting of the observed digital number (DN) values with GF-5/AHSI and predicted at-sensor radiances with MODTRAN 5 based on extended hyperspectral surface reflectance. Comparisons with onboard calibration results were also performed, and the averaged relative differences were within 5% with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1\delta $ </tex-math></inline-formula> standard deviations less than 10% for most bands, except for those in the atmospheric absorption and low signal-to-noise ratio bands. The comparison results indicate that the on-site radiometric calibration results are consistent with the onboard results, and the operational on-orbit radiometric calibration approach is reliable in the case that there are no measurements in the SWIR spectra range. The on-orbit radiometric performance of GF-5/AHSI rapidly degraded during the first several months after its launch and then tended to be relatively stable.

A Classification-Based, Semianalytical Approach for Estimating Water Clarity From a Hyperspectral Sensor Onboard the ZY1-02D Satellite

Water clarity (Zsd) is a widely used quality indicator that can be estimated from remote sensing imagery. China’s newest generation Advanced HyperSpectral Imager (AHSI) onboard the ZY1-02D satellite is expected to enable accurate water clarity retrieval for inland waters, since AHSI can provide abundant band choices while its 30-m spatial resolution is advantageous for monitoring small inland water bodies. In this study, to retrieve Zsd from the ZY1-02D imagery for inland waters with varying turbidities, we propose a classification-based, semi-analytical method in which the red/blue band ratio is employed to distinguish clear to moderately turbid water and highly turbid waters. Two Quasi Analytical Approaches (QAAs), QAAv5 and QAAm14, are used to estimate the total absorption coefficient (a(λ)) and the backscattering coefficient (bb(λ)) for clear to moderately turbid water and highly turbid waters, respectively. The estimated a(λ) and bb (λ) are utilized to obtain the diffuse attenuation coefficient Kd, followed by the Zsd calculations. Compared with 70 matchups of in situ measured Zsd values (0–6.5 m), the ZY1-02D image-derived Zsd achieved an R2 of 0.98, with an average unbiased relative error and root mean square error of 29.1% and 0.52 m, respectively. In addition, the proposed method can yield Zsd with higher accuracies than that of optimized empirical models. Therefore, the ZY1-02D AHSI imagery can retrieve reliable Zsd for both clear (> 3 m) and turbid waters (0–3.0 m), thereby serving as a useful satellite data source for monitoring the water clarity of large-scale inland water bodies.

Vicarious Radiometric Calibration of the AHSI Instrument Onboard ZY1E on Dunhuang Radiometric Calibration Site

The Advanced Hyperspectral Imager (AHSI) is the second hyperspectral imager of China, which is also one of the most important payloads onboard the ZY1E satellite. In order to monitor the radiometric calibration status since its launch on 12 Sep. 2019, and thus provide supplementary on-board radiometric calibration, this paper conducts several vicarious radiometric calibration experiments for the ZY-1E AHSI. Five satellite observations over Dunhuang radiometric calibration site, one of the most relevant China Radiometric Calibration Sites (CRCS), were used. Surface reflectance, radiosonde data, aerosol optical depth (AOD) loading, and water vapor content were used to simulate the top of atmosphere (TOA) radiance at the entrance pupil of the satellite via MODerate resolution atmospheric TRANsmission (MODTRAN). Our results show that the vicarious radiometric calibration coefficients are relatively constant with the official coefficients: the mean relative differences are 5.55%, 5.64%, 7.02%, 5.07%, and 5.30% on 11 Jan, 17 Jan, 12 Feb, 07 May, and 23 Nov, 2021, respectively. The vicarious radiometric calibration coefficients were found in good agreement with the official ones according to the validation analysis based on different surface types, i.e., water and vegetation. Uncertainties of vicarious radiometric calibration due to AOD assumptions, AOD measurements, water vapor measurements, radiosonde data measurements, relative spectral response (RSR) shifts and surface reflectance measurements are also discussed in detail. The obtained results show that the ZY-1E AHSI exhibits good on-orbit radiometric status.