Method and system for geometric correction of UAV (unmanned aerial vehicle) loaded imaging hyperspectrum

Abstract

The invention relates to a method and a system for the geometric correction of UAV (unmanned aerial vehicle) loaded imaging hyperspectrum. The method comprises the steps of collecting the position and attitude information of the POS sensor with low precision of an existing UAV; analyzing the precise position and attitude information in the photographing centre of a numerical code photograph according to the position and attitude information, so as to generate a DEM (Dynamic Effect Model) of a photograph covering area; correcting position and attitude data corresponding to a plurality of imaging hyperspectral scanning lines in the photographing centre of an adjacent numerical code photograph according to the precise position and attitude information of the photographing centre, so as to obtain the linear array position and attitude information with high precision of the plurality of imaging hyperspectral scanning lines; building a collinearity equation according to the linear array position and attitude information with high precision and the DEM, so as to generate a hyperspectral image. The POS information with high precision is calculated by calculating the area array imaging data of the UAV, the POS data with low precision of the UAV can be optimized, the precise geometric correction of the scanning lines of an imaging hyperspectral imager can be realized, and the technical support is provided for the extensive application of UAV loaded imaging hyperspectrum.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method and system for hyperspectral geometric correction of UAV-borne imaging. Background technique [0002] POS (position and orientation system) positioning and orientation system, the POS sensor is used to collect position and attitude information. [0003] DEM (Digital Elevation Model) digital elevation model is a virtual representation of landforms. It can be obtained through photogrammetry based on aerial or satellite images. It is used as the basic data for digital image orthorectification. necessary condition. [0004] In recent years, imaging hyperspectral remote sensing technology has promoted the quantitative development of agricultural remote sensing with the characteristics of map-spectrum integration and high spatial and spectral resolution, and has great application potential. However, the satellite remote sensing platform is limited by the orbit, and ...

AI Technical Summary

Problems solved by technology

[0007] The technical problem to be solved by the present invention is: the UAV platform has poor stability and is easily affected by airflow. UAV-borne imaging hyperspectral usually adopts line array sensor push-broom, and the position and posture of the line array change significantly during the imaging process, resulting in Serious geometric distortion; UAV-borne imaging hyperspectral single line array coverage is small, feature extraction is difficult, the overlap between line arrays is low, adjacent line array feature matching is difficult, and the DEM of the survey area cannot be effectively extracted; and there is no The human-machine load capacity is low, and it is usually impossible to carry a high-precision POS system to obtain high-precision imaging hyperspectral line array position and attitude information

Images