A terrain segmentation network for navigable areas with global strip reliability evaluation and dynamic fusion

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: unknown, P. 125964 - 125964

Published: Dec. 1, 2024

Language: Английский

Deep Learning Method for Wetland Segmentation in Unmanned Aerial Vehicle Multispectral Imagery

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(24), P. 4777 - 4777

Published: Dec. 21, 2024

This study highlights the importance of unmanned aerial vehicle (UAV) multispectral (MS) imagery for accurate delineation and analysis wetland ecosystems, which is crucial their conservation management. We present an enhanced semantic segmentation algorithm designed UAV MS imagery, incorporates thermal infrared (TIR) data to improve outcomes. Our approach, involving meticulous image preprocessing, customized network architecture, iterative training procedures, aims refine boundary delineation. The demonstrates strong results, including a mean pixel accuracy (MPA) 90.35% intersection over union (MIOU) 73.87% across different classes, with (PA) 95.42% (IOU) 90.46% class. integration TIR not only enriches feature set but also, some extent, helps address imbalance issues, contributing more refined ecological analysis. along development comprehensive dataset that reflects diversity environments advances utility remote sensing technologies in monitoring. research lays groundwork detailed informative UAV-based evaluations health integrity.

Language: Английский

Few-Shot Learning with Adaptive Weight Masking in Conditional GANs

2022 International Conference on Electronics and Devices, Computational Science (ICEDCS), Journal Year: 2024, Volume and Issue: unknown, P. 435 - 439

Published: Sept. 23, 2024

Language: Английский