Leaf area index (LAI) is an essential vegetation parameter that represents the light energy utilization and vegetation canopy structure. As the only in-operation hyperspectral satellite launched by China, GF-5 is potentially useful for accurate LAI estimation. However, there is no research focus on evaluating GF-5 data for LAI estimation. Hyperspectral remote sensing data contains abundant information about the reflective characteristics of vegetation canopies, but these abound data also easily result in a dimensionality curse. Therefore, feature selection (FS) is necessary to reduce data redundancy to achieve more reliable estimations. Currently, machine learning (ML) algorithms have been widely used for FS. Moreover, the same ML algorithm is usually conducted for both FS and regression in LAI estimation. However, no evidence suggests that this is the optimal solution. Therefore, this study focuses on evaluating the capacity of GF-5 spectral reflectance for estimating LAI and the performances of different combination of FS and ML algorithms. Firstly, the PROSAIL model, which coupled leaf optical properties model PROSPECT and the scattering by arbitrarily inclined leaves (SAIL) model, was used to generate simulated GF-5 reflectance data under different vegetation and soil conditions, and then three FS methods, including random forest (RF), K-means clustering (K-means) and mean impact value (MIV), and three ML algorithms, including random forest regression (RFR), back propagation neural network (BPNN) and K-nearest neighbor (KNN) were used to develop nine LAI estimation models. The FS process was conducted twice using different strategies: Firstly, three FS methods were conducted to search the lowest dimension number, which maintained the estimation accuracy of all bands. Then, the sequential backward selection (SBS) method was used to eliminate the bands having minimal impact on LAI estimation accuracy. Finally, three best estimation models were selected and evaluated using reference LAI. The results showed that although the RF_RFR model (RF used for feature selection and RFR used for regression) achieved reliable LAI estimates (coefficient of determination (R²) = 0.828, root mean square error (RMSE) = 0.839), the poor performance (R² = 0.763, RMSE = 0.987) of the MIV_BPNN model (MIV used for feature selection and BPNN used for regression) suggested using feature selection and regression conducted by the same ML algorithm could not always ensure an optimal estimation. Moreover, RF selection preserved the most informative bands for LAI estimation so that each ML regression method could achieve satisfactory estimation results. Finally, the results indicated that the RF_KNN model (RF used as feature selection and KNN used for regression) with seven GF-5 spectral band reflectance achieved the better estimation results than others when validated by simulated data (R² = 0.834, RMSE = 0.824) and actual reference LAI (R² = 0.659, RMSE = 0.697).

Convolutional neural networks (CNNs) have been widely applied in hyperspectral imagery (HSI) classification. However, their classification performance might be limited by the scarcity of labeled data to be used for training and validation. In this paper, we propose a novel lightweight shuffled group convolutional neural network (abbreviated as SG-CNN) to achieve efficient training with a limited training dataset in HSI classification. SG-CNN consists of SG conv units that employ conventional and atrous convolution in different groups, followed by channel shuffle operation and shortcut connection. In this way, SG-CNNs have less trainable parameters, whilst they can still be accurately and efficiently trained with fewer labeled samples. Transfer learning between different HSI datasets is also applied on the SG-CNN to further improve the classification accuracy. To evaluate the effectiveness of SG-CNNs for HSI classification, experiments have been conducted on three public HSI datasets pretrained on HSIs from different sensors. SG-CNNs with different levels of complexity were tested, and their classification results were compared with fine-tuned ShuffleNet2, ResNeXt, and their original counterparts. The experimental results demonstrate that SG-CNNs can achieve competitive classification performance when the amount of labeled data for training is poor, as well as efficiently providing satisfying classification results.

Water clarity, commonly measured as the Secchi disk depth ( $Z_{sd}$ ), is an important parameter that depicts water quality in aquatic ecosystems. China’s new generation Advanced HyperSpectral Imager (AHSI) on board the GF-5 satellite has significant potential for applications of more accurate water clarity estimation compared with existing multispectral satellite imagery, considering its high spectral resolution with a 30-m spatial resolution. In this study, we validate the semi-analytical model with various Quasi-Analytical Algorithms (QAA), including $QA{A}_{V5}$ , $QA{A}_{V6}$ , $QA{A}_{L09}$ and $QA{A}_{M14}$ , for the AHSI images with concurrent in situ measurements in four inland water bodies with a $Z_{sd}$ range of 0.3–4.5 m. The semi-analytical method with $QA{A}_{V5}$ can yield the most accurate $Z_{sd}$ predictions with approximated atmospheric-corrected remote sensing reflectance. For 84 concurrent sampling sites, the estimated $Z_{sd}$ had a mean absolute error (MAE) of 0.35 m, while the mean relative error (MRE) was 25.3%. Specifically, the MAEs of estimated $Z_{sd}$ were 0.22, 0.46, and 0.24 m for $Z_{sd}$ of 0.3–1, 1–3, and 3–4.5 m, respectively. The corresponding MREs were 33.1%, 29.1% and 6.3%, respectively. Although further validation is still required, especially in terms of highly turbid waters, this study indicates that AHSI is effective for water clarity monitoring.

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 ($\rho$) and satellite data quality, and applied to SOMC level mapping and estimation. The results show that: (1) The SI with the highest $\rho$ 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.

In this study, an on-orbit radiometric calibration campaign of the GF5-02 AHSI was performed at the RadCalNet Baotou site, based on the automated observation of reflectance and atmospheric parameters of a 300 m × 300 m homogeneous desert area. The consistency of the radiometric calibration coefficients was validated both at the Dunhuang calibration site and the Baotou site. The average relative difference between the calibrated top-of-atmospheric (TOA) radiance and the predicted TOA radiance were less than 7%. The R² of these two TOA radiances were all higher than 0.99. These results showed that the accuracy of calibration coefficients could meet the requirements of hyperspectral quantification applications. The uncertainty of GF5-02 AHSI radiometric calibration was 6.18%. This study also demonstrated that automated observation data of the Baotou site were reliable for high-frequency radiometric calibration and radiometric performance monitoring of GF5-02 AHSI.

The main feature of grassland degradation is the change in the vegetation community structure. Hyperspectral-based grassland community identification is the basis and a prerequisite for large-area high-precision grassland degradation monitoring and management. To obtain the distribution pattern of grassland communities in Xilinhot, Inner Mongolia Autonomous Region, China, we propose a systematic classification method (SCM) for hyperspectral grassland community identification using China’s ZiYuan 1-02D (ZY1-02D) satellite. First, the sample label data were selected from the field-collected samples, vegetation map data, and function zoning data for the Nature Reserve. Second, the spatial features of the images were extracted using extended morphological profiles (EMPs) based on the reduced dimensionality of principal component analysis (PCA). Then, they were input into the random forest (RF) classifier to obtain the preclassification results for grassland communities. Finally, to reduce the influence of salt-and-pepper noise, the label similarity probability filter (LSPF) method was used for postclassification processing, and the RF was again used to obtain the final classification results. The results showed that, compared with the other seven (e.g., SVM, RF, 3D-CNN) methods, the SCM obtained the optimal classification results with an overall classification accuracy (OCA) of 94.56%. In addition, the mapping results of the SCM showed its ability to accurately identify various ground objects in large-scale grassland community scenes.