Discriminant Analysis of the Damage Degree Caused by Pine Shoot Beetle to Yunnan Pine Using UAV-Based Hyperspectral Images

提出了一种低成本、小型化的基于无人机的高光谱图像数据分析系统。我们的研究显示了基于无人机高光谱遥感技术用于森林病虫害监测的潜力。这种技术可以有效地在高精度的遥感数据上对树木异常反射率特征进行小区域判别。我们通过Pearson相关分析和逐步判别分析分析小蠹虫对云南松造成的损伤，得到其敏感波段和光谱参数。首先验证地面调查枝梢被害率与冠层被害率的关系，发现二者呈现极显著正相关。然后采用手绘ROI提取树冠光谱，分析云南松受害光谱特征。然后，实现了敏感波段和敏感光谱参数的选择。基于敏感波段和参数建立小蠹虫危害程度监测模型，并建立定量化判定规则。

A low-cost, miniaturized unmanned aerial vehicle (UAV)-based hyperspectral image data analysis system is proposed. Our research demonstrates the potential of UAV-based hyperspectral remote sensing technology for forest pest and disease monitoring. This technology can effectively identify abnormal reflectance characteristics of trees in small areas using high-precision remote sensing data. We analyzed the damage inflicted by scolytid beetles on Yunnan pine via Pearson correlation analysis and stepwise discriminant analysis, and obtained the sensitive bands and spectral parameters associated with this damage. First, we verified the relationship between the branch tip infestation rate and the canopy infestation rate from field surveys, and found that the two exhibited an extremely significant positive correlation. Then, we manually delineated ROIs to extract canopy spectra and analyzed the spectral characteristics of damaged Yunnan pine. Subsequently, we selected the sensitive bands and spectral parameters. Based on the sensitive bands and parameters, we established a monitoring model for the damage severity of scolytid beetles, and formulated quantitative determination rules.

搜集汇总

背景与挑战

背景概述

该数据集利用无人机高光谱图像技术，研究小蠹虫对云南松的损害程度，通过光谱分析建立损害监测模型。数据属于人工林灾害防控研究项目，采用协议共享方式提供。