Few-shot hyperspectral classification is a challenging problem that involves obtaining effective spatial–spectral features in an unsupervised or semi-supervised manner. In recent years, as a result of the development of computer vision, deep learning techniques have demonstrated their superiority in tackling the problems of hyperspectral unmixing (HU) and classification. In this paper, we present a new semi-supervised pipeline for few-shot hyperspectral classification, where endmember abundance maps obtained by HU are treated as latent features for classification. A cube-based attention 3D convolutional autoencoder network (CACAE), is applied to extract spectral–spatial features. In addition, an attention approach is used to improve the accuracy of abundance estimation by extracting the diagnostic spectral features associated with the given endmember more effectively. The endmember abundance estimated by the proposed model outperforms other convolutional neural networks (CNNs) with respect to the root mean square error (RMSE) and abundance spectral angle distance (ASAD). Classification experiments are performed on real hyperspectral datasets and compared to several supervised and semi-supervised models. The experimental findings demonstrate that the proposed approach has promising potential for hyperspectral feature extraction and has better performance relative to CNN-based supervised classification under small-sample conditions.

High-resolution remote-sensing images can be used in human activity analysis and criminal activity monitoring, especially in sparsely populated zones. In this paper, we explore the applicability of China’s Gaofen satellite images in the land cover classification of Xinjiang, China. First of all, the features of spectral reflectance and a normalized radar cross section (NRCS) for different types of land covers were analyzed. Moreover, the seasonal variation of the NRCS in SAR (Synthetic Aperture Radar) images for the study area, Dunkuotan Village of Yuli County, China, was demonstrated by the GEE (Google Earth Engine) platform accordingly. Finally, the CART (classification and regression trees) algorithm of a DT (decision tree) was applied to investigate the classification of land cover in the western area of China when both optical and SAR images were employed. An overall classification accuracy of 83.15% with a kappa coefficient of 0.803 was observed by using GF-2/GF-3 images (2017–2021) in the study area. The DT-based classification procedure proposed in this investigation proved that Gaofen series remote-sensing images can be engaged to effectively promote the routine workflow of the administrative department.

This paper examines the output characteristics of the photovoltaic (PV) array in a laser wireless power transmission (LWPT) system considering atmospheric turbulence and pointing error r (r, θ). Firstly, we propose a theoretical model for studying the laser energy of each PV cell in non-Kolmogorov turbulence by introducing power in bucket (PIB). Then, to obtain the output characteristics of the PV array, we build the equivalent model based on the carrier continuity equation and bipolar transport equation. The results demonstrate that the power at the maximum power point (MPP) P _mpp and the system efficiency η ₂ descend 50.9% and 51.5%, in the cases of r (20mm, 45°) compared with r (0, 0), respectively, when the transmission distance is 1000 m. The variation range (5.1 V to 22.3 V) of the voltage at MPP V _mpp with r (10 mm, 45°) is distinctly larger than that (21.9 V to 23.2 V) with r (0, 0) when the transmission distance varies from 0 m to 2000 m. The findings of this study are valuable for the design of DC-DC converter, maximum power point tracking (MPPT) algorithm and the closed loop control circuit in LWPT system to make it efficient and stable operation.

Scientific balloons provide an inexpensive and reliable platform for near-space scientific experiments. The analysis of the balloon geometry and forces has always been a major concern for balloon designers. Most previous studies have focused solely on the fully inflated shapes and forces of balloons, analyzing only the membrane structure and simplifying the effects of internal and external gases into a gradient pressure difference. This approach lacks consideration of the fluid–structure interaction (FSI) of scientific balloons. This paper utilizes the Coupled Eulerian–Lagrangian (CEL) method in the Abaqus/Explicit simulation environment to analyze the FSI effects of scientific balloons under the influence of internal helium and external air. Three typical working conditions of scientific balloons are selected for simulation analysis. First, a three-dimensional spherical balloon is simulated during the ascent process to verify the correctness of the CEL simulation framework. This also demonstrates the membrane folding characteristics during balloon ascent, which could not be calculated in previous two-dimensional axisymmetric simulations. Next, the study explores balloon shapes that deviate from quasi-static pressure distributions due to the motion of internal helium. These include the “mushroom” shape observed during the dynamic launching of the balloon on the ground and the “sail” shape caused by lateral airflow. The mushroom shape arises from the sudden loss of the bottom constraint, causing the internal helium to move upward while being resisted by the air at the balloon’s top. The simulation successfully replicates the rapid waist transition and the downward concavity at the top due to air resistance, while also providing the corresponding force distribution. For the sail-shaped condition, the simulation analyzes the balloon’s tilt angle and the characteristic upturn of its windward surface. By a comparison with no-wind conditions, this study quantifies the impact of wind on the forces acting on the balloon, offering practical guidance for balloon launching. The CEL simulation framework established in this study not only provides a new tool for the FSI analysis of scientific balloons but also enriches the mechanical analysis results of these balloons.

In laser wireless power transmission (LWPT) system, the dynamically varying laser intensity or load on the photovoltaic (PV) cell may result in significant decrease of the power, efficiency and stability. The explicit steady and dynamic I-V characteristic equations are proposed innovatively to study the dynamic output characteristics and the stability of the PV cell under laser intensity or equivalent load disturbance. The time-domain response equations of photocurrent and dark current under large signal step disturbance are deduced. And the frequency-domain small signal model of the PV cell is derived to calculated the output impedance. The results suggest that the dynamic response time and the steady-state current of the photocurrent increase with the carrier lifetime. The PV cell under the high-frequency disturbance may result in the instability of the LWPT system. The models and methods will provide basis for the simulation and optimization of the DC-DC converter and the closed loop controller to improve the dynamic characteristics and stability of the LWPT system.