Long-term degradation of black soil has led to reductions in soil fertility and ecological service functions, which have seriously threatened national food security and regional ecological security. This study is motivated by the UN’s Sustainable Development Goal (SDG) 2—Zero Hunger, specifically, SDG 2.4 Sustainable Food Production Systems. The aim was to monitor the soil organic matter (SOM) content of black soil and its dynamics via hyperspectral remote sensing inversion. This is of great significance to the effective utilization and sustainable development of black soil resources. Taking the typical black soil area of Northeast China as an example, the hyperspectral data of ground features were compared with SOM contents measured in soil samples to correlate SOM with spectral features. Based on their quantitative relationship, a dynamic fitness inertia weighted particle swarm optimization (DPSO) algorithm is proposed, which balances the global and local search abilities of a particle swarm optimization algorithm. The DPSO algorithm is applied to the parameter adjustment of an artificial neural network (BPNN), which is used instead of a traditional error back propagation algorithm, to build a DPSO-BPNN model. Then a global optimal analytical expression of hyperspectral inversion is obtained to improve the generalization ability and stability of the remote sensing quantitative inversion model. The results show that DPSO-BPNN model is more stable and accurate than existing models, such as multiple stepwise regression, partial least squares, and BP neural network models (adjust complex coefficient of determination = 0.89, root mean square error = 1.58, relative recent deviation = 2.93). The results of DPSO-BPNN inversion are basically consistent with the trend in SOM contents measured during surface geochemical exploration. As such, this study provides a basis for hyperspectral remote sensing inversion and monitoring of the SOM contents in black soil.

Xiong’an New Area is defined as the future city of China, and the regulation of water resources is an important part of the scientific development of the city. Baiyang Lake, the main supplying water for the city, is selected as the study area, and the water quality extraction of four typical river sections is taken as the research objective. The GaiaSky-mini2-VN hyperspectral imaging system was executed on the UAV to obtain the river hyperspectral data for four winter periods. Synchronously, water samples of COD, PI, AN, TP, and TN were collected on the ground, and the in situ data under the same coordinate were obtained. A total of 2 algorithms of band difference and band ratio are established, and the relatively optimal model is obtained based on 18 spectral transformations. The conclusion of the strength of water quality parameters’ content along the four regions is obtained. This study revealed four types of river self-purification, namely, uniform type, enhanced type, jitter type, and weakened type, which provided the scientific basis for water source traceability evaluation, water pollution source area analysis, and water environment comprehensive treatment.

Keywords: Hyperspectral Data, Classification, Three-dimensional Convolution, Deep CNN, Feature Extraction Spatial and spectral information are obtained simultaneously by hyperspectral remote sensing. Joint extraction of these information of hyperspectral image is one of most import methods for hyperspectral image classification. In this paper, a novel deep convolutional neural network (CNN) is proposed, which extracts spectral-spatial information of hyperspectral images correctly. The proposed model not only learns sufficient knowledge from the limited number of samples, but also has powerful generalization ability. The proposed framework based on three-dimensional convolution can extract spectral-spatial features of labeled samples effectively. Though CNN has shown its robustness to distortion, it cannot extract features of different scales through the traditional pooling layer that only have one size of pooling window. Hence, spatial pyramid pooling (SPP) is introduced into three-dimensional local convolutional filters for hyperspectral classification. Experimental results with a widely used hyperspectral remote sensing dataset show that the proposed model provides competitive performance.

Mangrove is the key vegetation in the transitional zone between land and sea, and its health assessment can indicate the deep-level ecological information. The LAI and six key nutrients of mangrove were selected as quantitative evaluation indicators, and the decisive evaluation method of mangrove growth was expected. The mangrove reserve of Dongzhai Port National Nature Reserve in Hainan Province, China, was selected as the study area, with an area of 17.71 km². The study area was divided into adjacent urban areas, aquaculture areas, and agricultural production areas, and key indicators are extracted from satellite hyperspectral data. The extraction process includes spectral data preprocessing, spectral transformation, spectral combination, spectral modeling, and precision inspection. The spatial distribution of LAI and six key nutrient components of mangrove in the study area were obtained. LAI and Chla need to calculate the index after high-order differentiation of the spectrum; MSTR and Chlb need to calculate the envelope after the second-order differential of the spectrum; TN and TP are directly changed by original or exponential spectrum; the spectral transformation method adopted by TK was homogenization after first-order differential. The results of the strength of nutrient content along the three regions show that there was no significant difference in the retrieval index of mangroves in the three regions, and the overall health level of mangroves was consistent. Chla was the key identification component of mangrove growth and health. The contents of nutrient elements with correlation coefficient exceeding 0.80 include MSTR and TK (0.98), Chla and TP (0.96), Chla and TK (0.87), MSTR and Chla (0.86), MSTR and TK (0.83), and MSTR and TP (0.81). The study quantifies the relationship between different LAI and nutrient content of mangrove leaves from the perspectives of water, leaf biology, and chemical elements, which improved our understanding of the relationship between key components during mangrove growth for the first time.

This study innovatively leveraged proximal remote sensing to address the challenge of mineral exploration in vegetation-covered regions. Remote and proximal sensing has proven to be highly effective in pinpointing surface-exposed alteration minerals and detecting potential mining sites in previously unproductive areas. However, in regions where vegetation is abundant, the presence of foliage poses a significant challenge to mineral exploration efforts, creating a natural barrier that hinders the search for valuable minerals. In this study, we explored the linear relationship between the spectral changes induced by metals (specifically Fe and Mo) in wheat plants and the concentrations of these metal elements in different parts of the plant canopy at various growth stages. This investigation was conducted through meticulously designed controlled experiments to understand the interaction between metal elements in the soil and wheat plants. We have established linear models linking wheat biochemistry, vegetation spectroscopy, and soil concentration gradients of Fe and Mo. The analysis of Fe and Mo concentrations in leaves and wheat spikes across varying soil concentration gradients revealed significant positive correlations between the canopy accumulation sites and soil element concentrations (p < 0.05), with a correlation coefficient (R) exceeding 0.85, affirming the representativeness of these two canopy sites for subsequent spectral analysis and modeling. Regarding the wheat spectral analysis, the absorption features at specified wavelengths were identified as significant for creating valid linear models to analyze the effect of Fe and Mo in wheat leaf and spike spectra. Comparing the univariate (URL) and multivariate (MLR) models demonstrated that MLR modeling, incorporating multiple absorption feature parameters, provided more accurate results compared to scenarios with only one absorption feature in the modeling process (MLR: Fe-leaf: R² = 0.941, RMSE = 1.171; Mo-spike: R² = 0.934, RMSE = 0.042). To conclude, this study introduces a novel method that exploits the wheat spectral properties observed across different canopy sections during various growth stages of vegetation and under varying concentrations of Fe and Mo gradients. The methodology elucidated in this research provides technical support and lays the theoretical foundation for evaluating mineral resources in vegetated areas.

A novel method for detecting ships which aim to make full use of both the spatial and spectral information from hyperspectral images is proposed. Firstly, the band which is high signal-noise ratio in the range of near infrared or short-wave infrared spectrum, is used to segment land and sea on Otsu threshold segmentation method. Secondly, multiple features that include spectral and texture features are extracted from hyperspectral images. Principal components analysis (PCA) is used to extract spectral features, the Grey Level Co-occurrence Matrix (GLCM) is used to extract texture features. Finally, Random Forest (RF) model is introduced to detect ships based on the extracted features. To illustrate the effectiveness of the method, we carry out experiments over the EO-1 data by comparing single feature and different multiple features. Compared with the traditional single feature method and Support Vector Machine (SVM) model, the proposed method can stably achieve the target detection of ships under complex background and can effectively improve the detection accuracy of ships.