首页|冠层尺度的无人机高光谱染病单木识别

冠层尺度的无人机高光谱染病单木识别

扫码查看
基于马尾松感染松材线虫病后的光谱与纹理特征,探索机载高光谱影像在松材线虫病单木识别的应用效果与前景.本文分析了机载高光谱影像上不同程度染病与健康马尾松冠层的光谱特性,根据不同染病程度的光谱与纹理特征建立机载高光谱监督分类识别模型,并对不同监督分类方法识别结果进行分析与评价,提出适用于广东省的机载高光谱影像松材线虫病识别分类模型.研究表明,基于机载高光谱影像的松树冠层光谱曲线与地面非成像高光谱变化基本一致,这说明基于机载高光谱影像分析松树冠层光谱特性是可行的.同时,相较于光谱指数分析方法及高分辨率影像数据,基于高光谱影像的监督分类方法识别枯死松树可最大限度剔除裸土和阴影的干扰,但针对裸土边缘的干扰仍较难剔除.
Canopy-scale UAV Hyperspectral Diseased Single Tree Identification
Based on the spectral and texture characteristics of Pinus massoniana infected with Bursaphelenchus xylophilus,the applica-tion effect and prospect of airborne hyperspectral images in the identification of single tree infected with Bursaphelenchus xylophilus were explored. In this paper,the spectral characteristics of different degrees of diseased and healthy Pinus massoniana canopy on air-borne hyperspectral images were analyzed,and an airborne hyperspectral supervised classification and recognition model was estab-lished according to the spectral and texture characteristics of different degrees of disease,and the identification results of different su-pervised classification methods were analyzed. Based on the analysis and evaluation,an airborne hyperspectral image diseased Pinus massoniana identification and classification model suitable for Guangdong was proposed. The research shows that the spectral curve of pine canopy based on airborne hyperspectral images is basically consistent with the ground non-imaging hyperspectral changes,which indicates that it is feasible to analyze the spectral characteristics of pine canopy based on airborne hyperspectral images. At the same time,compared with the spectral index analysis method and high-resolution image,the supervised classification method based on hy-perspectral images can eliminate the interference of bare soil and shadow to the greatest extent,but it is still difficult to eliminate the interference of the edge of bare soil.

airborne hyperspectralcanopyBursaphelenchus xylophilussupervised classification

黄小川、王斌、吴志军、夏进亮、李琴

展开 >

广东省国土资源测绘院,广东广州 510663

自然资源部华南热带亚热带自然资源监测重点实验室,广东广州 510663

广东省自然资源科技协同创新中心,广东广州 510663

广东省测绘技术有限公司,广东广州 510663

展开 >

机载高光谱 冠层 松材线虫病 监督分类

2024

测绘与空间地理信息
黑龙江省测绘学会

测绘与空间地理信息

影响因子:0.788
ISSN:1672-5867
年,卷(期):2024.47(9)