在退化的沙漠草原中评估植物多样性指数,使用无人机超频谱多式联络数据和编码器-CNN
Zhaohui Tang1,2, Chuanzhong Xuan3,4, Tao Zhang1
1College of Mechanical and Electrical Engineering, Inner Mongolia Agricultural University, No. 306 Zhaowuda Road, Saihan District, Hohhot, 010018, Inner Mongolia, China.
Scientific reports
|August 20, 2025
概括
准确的沙漠草原植物多样性评估对于恢复至关重要. 这项研究使用多式无人机超光谱数据和深度学习来精确地绘制植物多样性,改善生态保护工作.
科学领域:
- 生态生态学 生态生态学
- 遥感 遥感 遥感 遥感
- 计算机科学 计算机科学
背景情况:
- 沙漠草原生态系统面临着气候变化和人类活动带来的生物多样性挑战.
- 准确的植物多样性评估对于有效恢复这些脆弱环境至关重要.
- 沙漠草原稀少的植被使传统的远程探测方法复杂化,以评估多样性.
研究的目的:
- 开发一种用于评估退化沙漠草原植物多样性指数的新方法.
- 提高植物多样性评估的准确性和效率,使用无人机 (UAV) 的高光谱数据.
- 融合多式联运特征,以提高分类和多样性指数的计算.
主要方法:
- 利用多式无人机超频谱数据,整合空间光谱,植被指数和纹理特征.
- 开发了一种新的编码器-CNN模型,将通道注意力融合 (CAF) 纳入跨层残余融合.
- 通过融合Encoder和CNN模型来计算像素级别的多样性指数,构建了一个全球-本地协同表达结构.
主要成果:
- 融合多式联网数据和深度学习方法准确地确定了植被类型和多样性指数,与地面真相数据保持一致.
- 拟议的方法在稀疏植被分类方面取得了很高的准确性 (总体为90.01%,平均为85.23%).
- 这种方法优于单模式分析和传统模型,如3DCNN和视觉变压器 (ViT).
结论:
- 多模无人机超谱数据与深度学习相结合,为评估沙漠草原植物多样性提供了强大的工具.
- 开发的编码器-CNN模型提供了准确的社区组成信息,这对于生态恢复至关重要.
- 这项研究强调了先进遥感和人工智能在生态保护和管理方面的潜力.
更多相关视频
相关概念视频
02:16Light Acquisition
8.6K
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
8.6K
02:22Introduction to Plant Diversity
45.8K
From Water to Land
45.8K
02:31Habitat Fragmentation
17.9K
Habitat fragmentation describes the division of a more extensive, continuous habitat into smaller, discontinuous areas. Human activities such as land conversion, as well as slower geological processes leading to changes in the physical environment, are the two leading causes of habitat fragmentation. The fragmentation process typically follows the same steps: perforation, dissection, fragmentation, shrinkage, and attrition.
17.9K