As a crucial crop for oil and protein, soybean is highly sensitive to soil moisture during growth. Therefore, it is crucial to accurately monitor soil moisture over soybean fields during the growth period. In the L-band Microwave Emission of the Biosphere (L-MEB) model, the calculation of vegetation transmissivity may have some limitations, potentially affecting soil moisture retrieval accuracy. To improve the effectiveness of the retrieval, we introduced the improved Beer-Lambert law to describe the attenuation effect of the vegetation layer on the microwave signals more accurately, thus improving the L-MEB model. The experimental results indicate that our improved method significantly enhances the accuracy of soil moisture retrieval over soybean fields: the correlation coefficients of soil moisture results retrieved using vegetation water content and leaf area index improved by 0.9% and 9.5%, respectively, and the root mean square errors decreased by 39.78% and 19.35%, respectively.

Enhanced sensor technology and advanced control algorithms have expanded the use of autonomous Unmanned Aerial Vehicles (UAVs) in critical sectors such as power inspection. However, the limitations in positioning accuracy and onboard computational power constrain the robustness and versatility of UAV motion planning algorithms in outdoor environments. Therefore, we enhanced the existing generalized UAV obstacle avoidance pathfinding methods for application in electric power inspection, ensuring effective performance despite limited GPS signal quality and computational power. Initially, we delineate impassable areas by marking the non-collision space beneath obstacles at a specific height based on actual requirements. Next, environmental factors are integrated into the assessment of local target points, mitigating risks of UAVs encountering obstacles during trajectory planning. Finally, by discretizing the output B-spline trajectory with node fitness, we tightly couple the UAV's real-time position with the initiation of trajectory replanning. Experimental results confirm the method's robustness and efficiency.

Soil moisture is of great importance for regional hydrological studies such as agricultural management and drought prediction, but those applications usually require a high spatial resolution of 1-10 kilometers. To improve the spatial resolution of satellite soil moisture products, this paper constructs a soil moisture downscaling method in Apparent Thermal Inertia- Leaf Area Index (ATI-LAI) space using an improved automatic edge determination algorithm. The effectiveness and robustness of the method are validated by the Australian Soil Moisture Monitoring Network and ESA Climate Change Initiative (ESA CCI) soil moisture data. By comparing with in situ SM measurements and the spatial patterns of the CCI SM, the R of the downscaled 1km SM reaches 0.694 with a bias of 0.027. The results show that the downscaled 1km SM significantly improves the spatial details of the CCI SM while replicating the accuracy of the CCI SM, and demonstrating fine-scale spatial variability.

Clouds greatly affect the quality of optical remote-sensing image data. Algorithms for detecting and removing clouds can significantly enhance the utilization of optical data. Numerous studies highlight the crucial role of cirrus bands in cloud detection and removal, although only a few satellites possess this capability. In this research, a unified theory based on physical model is proposed and validated for thin cloud detection and removal. First, top of reflectance (TOA) values of thin clouds are detected in a specific band. Subsequently, the detected cloud image is used to remove thin clouds from other bands via spatial transformation. Experiments with actual Landsat-9 Operational Land Imager 2 (OLI-2) data confirm the effectiveness of the proposed approach both qualitatively and quantitatively. Even if thin clouds are undetectable in the cirrus bands, or cirrus bands are unavailable, this research still presents a novel paradigm for detecting and removing thin clouds.

To achieve high-quality inverse synthetic aperture radar (ISAR) imaging, measuring amplitude and phase precisely is crucial. However, for moving targets with complex motions such as rotation or rocking, the coherence between pulses deteriorates, making it difficult to estimate phase information accurately. This leads to azimuth defocusing issues with the traditional range-Doppler (RD) algorithm. To overcome this obstacle, we propose a novel ISAR imaging method based on the expectation maximization (EM) algorithm. This method accurately estimates the motion trajectory of the target by using only the envelope information of time-varying amplitudes and the dynamic observation model established within the Bayesian framework. The parameters are adjusted with great precision through iterative optimization of the EM algorithm, which includes both the expectation and maximization steps, to achieve high-quality ISAR images and track target trajectory effectively. The effectiveness of the proposed algorithm is verified using simulation data based on the millimeter-wave (MMW) radar on the Yak-42 aircraft model.

Leaf area index (LAI) is a critical plant biophysical parameter required for modelling plant photosynthesis and crop yield estimation. UAV remote sensing plays an increasingly significant role in providing the data source needed for LAI extraction. This study proposed a method that automatically calculate crop effective LAI (LAIe) using UAV-based 3-D point cloud. The porosity and projection function of crops from different zenith perspectives was estimated using three-dimensional perspective. Then the LAIe was calculated using the Beer Lambert law. The result shows a good linear correlation between the calculated LAIe and the field LAI measured by digital hemispherical photography method, and R2is 0.64. The method presented in this paper performs well in LAIe estimation of main leaf development stages of winter wheat growth period. It offers an effective means for mapping crop LAIe without reference data and saves time and cost.

Three dimensions (3D) reconstruction of synthetic aperture radar (SAR) has previously been achieved through SAR tomography or circular SAR techniques. To advance SAR 3D reconstruction methods and accommodate applications requiring high efficiency, this paper presents a novel 3D reconstruction method using single-channel and single-pass stripmap SAR images. The proposed method involves three main stages: dechirping image generation, SAR image matching, and 3D coordinates calculation. First, a dechirping domain SAR image is generated using the dechirping process (Dechirp-SAR) instead of matched filtering during the azimuth compression. Second, the local feature matching with Transformer (LoFTR) algorithm is utilized for SAR image matching, followed by filtering the results. Finally, the 3D coordinates are calculated using the range-Doppler (RD) equations within the dechirping domain (Dechirp-RD). Experiments on both simulated and measured data quantitatively validate the effectiveness of the proposed method.

Simultaneous Localization and Mapping (SLAM) is a technique employed to estimate the motion of an agent (or target) and reconstruct structures within unknown environments. Recently, the feasibility of SLAM utilizing millimeter wave (MMW) radar data in conjunction with other data types has been demonstrated within a visual framework. However, the fusion and processing of data from various sensors can result in increased costs and complex algorithms. In this paper, we propose an outdoor SLAM approach based solely on MMW data within a LiDAR framework, aiming to offer a cost-effective solution while enhancing robustness and preserving high accuracy. To address the problem of an inadequate sampling rate in MMW data, slicing and interpolating are employed to augment the dataset. A quantitative analysis of the proposed SLAM method using MMW data is conducted, and the results demonstrate its effectiveness, with performance closely approximating that of LiDAR-based SLAM.

Low-altitude unmanned aerial vehicle (UAV) remote sensing facilitates the frequent detection of power lines and liberates manual inspection. In the process of UAV power line inspection, power line detection in UAV aerial images plays an important role. But false alarms and miss alarms often occur during the detection process. Aiming at this problem, a power line detection method based on Maxtree is proposed. This method transforms the UAV aerial images into a graph structure, i.e., Maxtree, and detects the power lines under graph signal processing frame. Two-stage filtering is designed to preserve power line components. The preprocessing stage filters out most of the background part according to the color value, and the other stage performs filtering on the Maxtree created with connectivity and gray value. For each node, three attribute components, i.e., gray value, linearity, and length, are assigned to facilitate power line detection. Experiments show that the method can detect power lines accurately and effectively.

This study proposes a novel thin cloud removal algorithm based on sorted slow feature analysis (s-SFA), exploiting the spatial slow variation feature of thin clouds. The algorithm is applied to Landsat-8 Operational Land Imager (OLI) data. The effectiveness of the proposed algorithm is assessed using both simulated and real cloud-affected data. Results demonstrate significant improvements in R2and peak signal-to-noise ratio (PSNR) values for the simulated cloud-affected data, indicating efficient thin cloud removal. Further validation using real Landsat-8 data confirms the algorithm's effectiveness over various land use and land cover types while preserving satisfactory ground features in cloud-free regions.