Reporters recently learned from the Beijing Academy of Agriculture and Forestry Sciences that the National Agricultural Informatization Engineering Technology Research Center,the Information Technology Research Center of the Beijing Academy of Agriculture and Forestry Sciences,in conjunction with the Xi'an Institute of Optics and Precision Machinery of the Chinese Academy of Sciences,has successfully developed the first home-made new video hyperspectral and point cloud"figure spectrum integration"sensor.This sensor can efficiently and accurately capture the phenotype information of crops,providing strong technical support for crop breeding and precision agriculture.
  "This sensor adopts a modular design that integrates hyperspectral acquisition with 3D point cloud measurement.It has the characteristics of lightweight and agile,and is suitable for use in light and small unmanned aerial vehicles and ground mobile platforms.Through this sensor,researchers can quickly obtain millimeter resolution imaging hyperspectral data and synchronously obtain fine 3D crop morphology data,achieving real-time synchronization perception of crop spatial 3D,spectral dimension,and temporal dimension in five dimensions."Yang Guijun,director of the Key Laboratory of Agricultural Remote Sensing Mechanism and Quantitative Remote Sensing of the Ministry of Agriculture and Rural Affairs,told reporters.
  The reporter learned that in terms of technology,the sensor has made three significant breakthroughs."Firstly,it adopts domestically produced pixel level spectral filtering snapshot array spectroscopy technology to achieve video level high-speed synchronous imaging.This technology not only overcomes the problem of geometric distortion in traditional linear array push broom hyperspectral images,but also achieves the smallest volume in the industry for the integrated hyperspectral and lidar system,with weight controlled within 3 kilograms."Yang Guijun said.
  Secondly,the sensor integrates advanced domestic high-precision surveying grade LiDAR,adopts a unique three sided tower mirror scanning structure and high-frequency motor,and generates 200-300 laser profiles per second.Under the same flight conditions,its point density is 4-6 times that of traditional LiDAR,and its absolute spatial position accuracy is better than 3 centimeters.This technology can achieve precise three-dimensional phenotype analysis of crops such as stem and leaf distribution,plant height,etc.in the field,providing more accurate data support for crop breeding.
  "Finally,the sensor also achieves high-precision image spectral fusion of 3D point clouds and hyperspectral images on mobile platforms such as drones.The geometric measurement accuracy of this technology is better than one pixel,providing strong support for the phenotypic collaborative analysis of crop breeding from indoor to outdoor and from a single'morphological structure'to'functional structure'."Yang Guijun said.(Reporter Ma Aiping)