Unmanned aerial vehicles (UAVs) are becoming increasingly popular in various fields such as agriculture, forest protection, resource exploration, and so on, due to their ability to capture high-resolution images quickly and efficiently at low altitudes. However, real-time image mosaicking of UAV image sequences, especially during long multi-strip flights, remains challenging. In this paper, a real-time incremental UAV image mosaicking framework is proposed, which only uses the UAV image sequence, and does not rely on global positioning system (GPS), ground control points (CGPs), or other auxiliary information. Our framework aims to reduce spatial distortion, increase the speed of the operation in the mosaicking process, and output high-quality panorama. To achieve this goal, we employ several strategies. First, the framework estimates the approximate position of each newly added frame and selects keyframes to improve efficiency. Then, the matching relationship between keyframes and other frames is obtained by using the estimated position. After that, a new optimization method based on minimizing weighted reprojection errors is adopted to carry out precise position calculation of the current frame, so as to reduce the deformation caused by cumulative errors. Finally, the weighted partition fusion method based on the Laplacian pyramid is used to fuse and update the local image in real time to achieve the best mosaic result. We have carried out a series of experiments which show that our system can output high-quality panorama in real time. The proposed keyframe selection strategy and local optimization strategy can minimize cumulative errors, the image fusion strategy is highly robust, and it can effectively improve the panorama quality.

Unmanned aerial vehicle (UAV) remote sensing has been applied in various fields due to its rapid implementation ability and high-resolution imagery. Single-UAV remote sensing has low efficiency and struggles to meet the growing demands of complex aerial remote sensing tasks, posing challenges for practical applications. Using multiple UAVs or a UAV network for remote sensing applications can overcome the difficulties and provide large-scale ultra-high-resolution data rapidly. UAV network path planning is required for these important applications. However, few studies have investigated UAV network path planning for remote sensing observations, and existing methods have various problems in practical applications. This paper proposes an optimization algorithm for UAV network path planning based on the vehicle routing problem (VRP). The algorithm transforms the task assignment problem of the UAV network into a VRP and optimizes the task assignment result by minimizing the observation time of the UAV network. The optimized path plan prevents route crossings effectively. The accuracy and validity of the proposed algorithms were verified by simulations. Moreover, comparative experiments with different task allocation objectives further validated the applicability of the proposed algorithm for various remote sensing applications

There are two widely recognized global fields in nature: the gravity field and the geomagnetic field. Universal gravitation and Earth rotation are important sources of the Earth’s gravity and geomagnetic fields, which are well known to us. After years of long-term observation, global research, and analysis, it was discovered that we have neglected a direct incident energy of the universe on the Earth. Solar radiation, leading to energy exchange from the atmosphere 100 km above the land surface, is the energy source of the Earth. Polarization is one of the four basic physical properties of solar radiation. After the solar radiation reaches the surface of these media, it reflects, scatters or refracts, and exhibits different degrees of polarization. The polarized solar light forms the Earth–sky polarization vector field. The polarized light dispersion is expected to become a new method for global analysis of the human environment. Polarization detection is the best way to accurately explore the atmospheric effects. Local polarized skylight distribution was found in different sites in the world; however, the global distribution of the polarized sunlight radiation has never been explored. In this paper, we investigate the Global Skylight Polarization Field. This study aimed at providing new insight into the laws of polarization over our Earth. We use a Rayleigh scattering model to obtain the simulation results of the sky polarization field. Rayleigh scattering occurs when the particle size is much smaller than the wavelength of the incident electromagnetic wave. We also use a polarized fisheye camera to collect the sky polarization image and calculate the distribution pattern of the DOLP (degree of linear polarization) and AOLP (azimuth of linear polarization) of the skylight. The stability and gradual change in the degree of polarization in the zenith direction are verified, and the distribution law and daily change law of the degree of polarization in the sky are obtained. With the increase in the solar altitude angle, the degree of polarization will decrease. We also observed the skylight polarization in different weather conditions. Our results demonstrate the physical basis, characteristics, and usability of the polarization field. They show an inevitable trend from optical remote sensing to polarization remote sensing.

Water ecosystems are highly sensitive to environmental conditions, including meteorological factors, which influence dissolved oxygen (DO) concentrations, a critical indicator of water quality. However, the complex relationships between multiple meteorological factors from various sites and DO concentrations pose a significant challenge for accurate prediction. This study introduces an innovative framework for enhancing DO concentration predictions in water bodies by integrating multi-station meteorological data. We first construct a dynamic meteorological graph with station-specific factors as node features and geographic distances as edge weights. This graph is processed using a Geo-Contextual Graph Embedding Module, leveraging a Graph Convolutional Network (GCN) to distill geographical and meteorological features from multi-station data. Extracted features are encoded and then temporally merged with historical DO values to form time-series data. Finally, a Temporal Transformer module is used for future DO concentration predictions. The proposed model shows superior performance compared to traditional methods, successfully capturing the complex relationships between meteorological factors and DO levels. It provides an effective tool for environmental scientists and policymakers in water quality monitoring and management. This study suggests that the integration of graph-based learning and a Temporal Transformer in environmental modeling is a promising direction for future research.