Publications by authors named "Ruopu Ma"
InSAR monitoring technology is widely used in investigating landslide hazards. Leveraging object detection algorithms to quickly extract landslide information from Wide-Area InSAR measurements is of great significance. Our InSAR-YOLOv8, an algorithm that automatically detects landslides from InSAR measurements, addresses the low accuracy and suboptimal detection performance of existing network models.
View Article and Find Full Text PDFArticle Synopsis
- The study focuses on improving the instance segmentation of individual maize plants using multispectral remote sensing data from UAVs, which is essential for efficient crop monitoring and management.
- Six instance segmentation algorithms are evaluated, with YOLOv8 delivering impressive accuracy, particularly in the NRG band, achieving bbox_mAP50 and segm_mAP50 scores of 95.2% and 94%.
- The research also explores the effects of varying resolutions on segmentation accuracy, finding that YOLOv8 maintains high performance even at lower resolutions suitable for phenotypic analysis.