渐进的注意力增强的高效Net-UNet用于从卫星图像中强大的水体映射.
Mohamed Ezz1, Alaa S Alaerjan1, Ayman Mohamed Mostafa2
1Computer Science Department, College of Computer and Information Sciences, Jouf University, Sakaka 72388, Saudi Arabia.
这项研究引入了一种先进的深度学习模型,用于在卫星图像中准确检测水体. 这种新的方法增强了可持续的水资源管理和气候适应性基础设施的发展.
科学领域:
- 遥感 遥感 遥感 遥感
- 计算机视觉 计算机视觉
- 环境科学 环境科学
背景情况:
- 在卫星图像中准确识别水体对于可持续的水资源管理和气候适应性基础设施至关重要.
- 现有的细分方法经常与复杂的水体模式和边界作斗争.
研究的目的:
- 开发一种新的深度学习架构,用于从卫星数据中高保真地提取水体.
- 用注意力机制来提高水体细分的精度,灵敏度和整体准确性.
主要方法:
- 将卷积区注意模块 (CBAM) 集成到经过修改的EfficientNet-UNet骨干中.
- 严格的训练使用五倍交叉验证,动态测试时间增大和Lovász损失优化.
- 在使用精度,灵敏度,特异性,准确性,子得分和IOU等指标的独立测试集上进行评估.
主要成果:
- 拟议的模型实现了高性能指标:精度 (90.67%),灵敏度 (86.96%),特异性 (96.18%),准确性 (93.42%),子得分 (88.78%) 和IOU (79.82%).
- 在提取复杂的水体特征方面,与传统的细分管道相比显著改进.
- 验证了注意力机制的有效性,以详细划分水体边界.
结论:
- 注意引导的深度学习网络为高准确度的水体绘图提供了强大而高效的途径.
- 开发的模型具有计算效率,适用于大规模的水资源和生态系统监测.
- 这项研究为针对远程传感应用的CBAM提供了定制的UNet风格架构.
更多相关视频
08:53Using a Classroom-Based Deese Roediger McDermott Paradigm to Assess the Effects of Imagery on False Memories
Published on: November 14, 2018
08:45Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example
Published on: October 24, 2012
相关概念视频
Body Water Content and Fluid Compartments
Energy of a Satellite in a Circular Orbit
Tumor Progression
Colon cancer is one of the best-documented examples of tumor progression. Early mutation in the APC gene in colon cells causes a small growth on the colon wall called a polyp. With time, this polyp grows into a benign, pre-cancerous tumor. Further...
States of Water
Water freezes when the intermolecular forces are greater than the kinetic energy. Unlike most other substances, water is less dense in its solid state than in its liquid state. This is because each water molecule can form...
Circular Orbits and Critical Velocity for Satellites
Nicolaus Copernicus (1473-1543) first suggested that the Earth and all other planets orbit the Sun in...
Satellite Stem Cells and Muscular Dystrophy
