Search

Over 90 % of the world's rice is produced in the Asia–Pacific region. Synthetic-aperture radar (SAR) enables all-day and all-weather observations of rice distribution in tropical and subtropical regions. The complexity of rice cultivation patterns in tropical and subtropical regions makes it difficult to construct a representative data-relevant rice crop model, increasing the difficulty in extracting rice distributions from SAR data. To address this problem, a rice area mapping method for large regional tropical or subtropical areas based on time-series Sentinel-1 SAR data is proposed in this study. Based on the analysis of rice backscattering characteristics in mainland Southeast Asia, the combination of spatiotemporal statistical features with good generalization ability was selected and then input into the U-Net semantic segmentation model, combined with WorldCover data to reduce false alarms, finally the 20 m resolution rice area map of five countries in mainland Southeast Asia in 2019 was obtained. The proposed method achieved an accuracy of 92.20 % on the validation sample set, and the good agreement was obtained when comparing our rice area map with statistical data and other rice area maps at the national and provincial levels. The maximum coefficient of determination R2 was 0.93 at the national level and 0.97 at the provincial level. These results demonstrate the advantages of the proposed method in rice area mapping with complex cropping patterns and the reliability of the generated rice area maps. The 20 m annual paddy rice area map for mainland Southeast Asia is available at 10.5281/zenodo.7315076 (Sun et al., 2022b). [ABSTRACT FROM AUTHOR]

Underwater ghost imaging based on deep learning can effectively reduce the influence of forward scattering and back scattering of water. With the help of data-driven methods, high-quality results can be reconstructed. However, the training of the underwater ghost imaging requires enormous paired underwater datasets, which are difficult to obtain directly. Although the Cycle-GAN method solves the problem to some extent, the blurring degree of the fuzzy class of the paired underwater datasets generated by Cycle-GAN is relatively unitary. To solve this problem, a few-shot underwater image generative network method is proposed. Utilizing the proposed few-shot learning image generative method, the generated paired underwater datasets are better than those obtained by the Cycle-GAN method, especially under the condition of few real underwater datasets. In addition, to reconstruct high-quality results, an underwater deblurring ghost imaging method is proposed. The reconstruction method consists of two parts: reconstruction and deblurring. The experimental and simulation results show that the proposed reconstruction method has better performance in deblurring at a low sampling rate, compared with existing underwater ghost imaging methods based on deep learning. The proposed reconstruction method can effectively increase the clarity degree of the underwater reconstruction target at a low sampling rate and promotes the further applications of underwater ghost imaging. [ABSTRACT FROM AUTHOR]

• YMPL can accurately detect targets in fog and expand the detection range. • New data preprocessing for creating bitmaps from multiple pulses. • The data preprocessing can clearly demonstrate the position of the target in the fog. • The YOLO and preprocessing combo improve target recognition in harsh weather. As one of the essential sensing technologies for autonomous driving, Lidar has not been widely adopted due to the significant impact of foggy and hazy weather leading to inaccurate target detection and distance measurement. In this paper, a YOLO-based Multi-Pulse Lidar system (YMPL) is proposed for accurate target detection in foggy conditions. The system integrates multiple one-dimensional pulse detection courses into a two-dimensional image and utilizes the YOLO target recognition algorithm to identify real target echoes and measure the distance of the target. The simulation and experimental results demonstrate that the YMPL system effectively mitigates the interference of fog and noise on pulse detection. Thereby the detection probability improves and the detection range extends. The system also shows the excellent anti-jitter ability. Under the circumstance of a 40 % backscattering coefficient, the system achieves a mean absolute error (MAE) of only 0.013 m within the range of 45.5 m, significantly outperforming the traditional threshold detection and ResNet, SVD-CNN and VIT algorithm. This lays a solid foundation for the all-weather practical application of lidar. [ABSTRACT FROM AUTHOR]