高辉 王秋平 赵新刚 潘从元

摘 要：现代农业的发展要求对农作物进行精准管理，需要对大面积农作物光谱信息进行精确和实时快速的监测，以便大面积、快速、无损的获取作物的生长、发育、养分和水分等的空间分布信息，为变量施肥、浇灌、施药等田间管理提供基本信息，是实现数字化农业的基础。利用反射光谱探测作物信息，建立基于特定目标测试数据的统计模型来实现作物叶面积指数、氮素、水分含量和灾害的光谱反演，已经取得了一定的研究成果。但由于建立的统计模型主要是通过地物光谱信息和卫星遥感影像数据之间的结合，模型精度受到特定的试验条件（时间、地点、对象）、空间分辨率、图谱的特异性等限制，很难进一步提高。该报告提出利用无人机和成像光谱仪结合，研制出基于无人机的农业低空高光谱的新型遥感技术平台，可以灵活地获取到高分辨率图谱合一数据。结合卫星遥感，将会对不同区域、不同品种作物的监测更灵活、更精确。可以满足现代农业生产对高空间分辨率、高时间分辨率、高光谱分辨率遥感信息的需求，将为解决精准农业中信息获取中的瓶颈问题提供一种解决方案。通过该研究实施，在国内首次开展了基于无人机的低空农作物成像光谱遥感平台的研制和应用示范，为低成本、高效、灵活实时获取大面积、高空间分辨率的农作物信息奠定了一定基础。报告提出和发展优化了田间农作物遥感成像信息探测的多项关键技术。（1）旋翼无人机超视距自主飞行的控制技术，包括自主起飞/降落、自主航迹跟踪、自主容错、自主定位与优化导航的功能；（2）低空遥感应用载荷的高精度支撑技术，用于支撑成像光谱仪系统以获得高清晰光谱图像；（3）成像光谱仪的微型化技术及由此带来的光谱图像畸变的矫正方法，用于微型成像光谱仪；（4）提出并实施了无人机悬停状态时通过反射镜旋转，实现光谱数据的采集和矫正；（5）建立基于低空高光谱遥感信息的应用模型，得到农作物不同生育期的成像光谱仪遥感反演模型和相应灾害遥感反演模型，为精准农业的实际应用提供依据。同时还开发了计算机软件进行快速数据获取和存储，并可在短时间内完成测得光谱的拼接和数据处理。 本报告研究成果还可以应用于环境、食品等科研和企业单位，可以为农业科研、生产管理、农产品质量检测、食品检测及环境生态检测等提供更好的技术手段。

关键词：旋翼无人机 成像光谱仪 遥感 农业 光谱分析

Abstract：Modern agriculture needs the crop spectral information in a large area in time. The spatial resolved spectral information will give the status of crop growth， nutrient and water， which is the basic information for field management as fertilization， irrigation and spraying.So imaging spectral information is the base of digital agriculture. To exploit the spectral information， a spectral information processing model is established that is based on corps reflect spectrum， to find out the leaf area index， nitrogen， moisture and disease Several research results are obtained based on the model. At present， above data processing model is based on the ground spectral information and satellite remote-sensing spectral information， the accuracy of the model is limited by the specific experiment conditions （different time， different height and different spatial resolution）. To improve the accuracy of the data model. The low attitude hyperspectal remote sensing platform based on unmanned aerial vehicle （UAV） is proposed. The platform employs the unmanned aerial vehicle and the image spectrometer， which can be used to get high spatial resolution image and high spectral resolution information for a large area.Combined the UAV based hyperspectral information and the satellite based hyperspectral information， we can monitor crops information in different spatial resolution and flexible way. This platform may be practical used as the high spatial resolution， high spectral resolution spectral information tools for modern agriculture. This is one of the solutions for the bottleneck of the accurate agricultural information collection. In this project， an UAV based hyperspectral information collection platform is developed and evaluated. The platform may provide a low-cost， efficient， flexible and real-time hyperspectral information tool with high spatial resolution of crops spectral information. Several key techniques related to the UAV based hyperspectral sensing for field crops are presented.（1）the UAV control technologies（including UAV take-off and landing， flight track plan， independent fault-tolerant， self-positioning and optimized navigation functions）.（2）the high precision supporting platform for spectrometer.（3）the micro hyperspectrometer development.（4）keeping the hyperspectrometer in a fixed position in the sky， to get hypespectral information by scanning a mirror.（5）developed an application model which may be used on UAV based hyperspectral information. Above research may provide the basis for the practical application of precision agriculture. Sixth， fast hyperspectral data acquisition and storage. The results of this project can also be applied in other research fields as environmental monitoring and food industry.

Key Words：Unmanned Aerial Vehicles；Imaging Spectrometer；Remote Sensing；Agriculture；Spectral Analysis

阅读全文链接（需实名注册）：http：//www.nstrs.cn/xiangxiBG.aspx？id=39142&flag=1