基于无人机高光谱影像的精细地物分类的研究

Research on Refinement Terrain Surface Features Classification Based on Hyperspectral Image of UAV

对局部区域入侵有害物种薇甘菊的精准发现和防治是一个难点,装有新型传感器的轻型无人机为复杂环境的数据采集提供了一种低成本、小风险的解决方案。通过无人机高光谱采集高分辨率影像,运用最大似然和支持向量机分析方法,对薇甘菊与其他物种精准分类识别进行研究。试验结果表明:基于机器学习方法的薇甘菊和其他物种的特征识别精度达到90%以上,无人机高光谱遥感可为局部区域薇甘菊的快速获取、精准监测预警和防治提供高效手段。

It is difficult to identify and control the invasive species of Mikania micrantha in local area.The light UAV equipped with new sensors provides a low-cost and low-risk solution for data collection in complex environment.Through the UAV hyperspectral acquisition of high-resolution images,using maximum likelihood and support vector machine analysis method,the precise classification and recognition of Mikania micrantha and other species were studied.The experimental results show that the recognition accuracy of Mikania micrantha and other species based on the sample training of deep learning is more than 90%.The UAV remote sensing data can provide efficient means for the accurate monitoring,early warning and control of Mikania micrantha in local areas.