Research On Object Location Of High Inclination Aerial Camera Based On Digital Elevation Model

Aerial camera is an important optical remote sensing equipment to acquire a large range of ground remote sensing images.It has a wide shooting range and high imaging resolution,and is widely used in geological exploration,resource detection,disaster relief and military detection.In particular,the high inclination aerial camera using oblique imaging greatly improves the efficiency of aerial reconnaissance,which is of great significance for the observation of complex terrain and border areas.In addition to high-resolution target imaging,aerial cameras also need to be able to accurately locate ground targets.However,high geodetic error in some complex terrain areas will seriously affect the positioning accuracy.Therefore,this paper proposes a target positioning algorithm suitable for high-dip aerial cameras in complex terrain,which improves the positioning accuracy and efficiency.This article introduces the coordinate transformation principle and the development status of positioning algorithm for ground target positioning by aerial camera.In view of the problem of large positioning error and low positioning efficiency in areas with large geodetic height error,a digital elevation model-based elevation half-interpolation iterative algorithm is proposed for ground target positioning,and a hardware positioning module is designed.The effectiveness and efficiency of the localization algorithm in this paper are verified by the outfield flight data and aerial images.The specific work of this paper is as follows:(1)This paper introduces the development history of aerial camera,analyzes the application range and limitations of the existing target location algorithm,establishes the basic coordinate system required in the target location process by the homogeneous coordinate transformation theory,deduces the transformation equation between different coordinate systems,realizes the target location process by collinear theory combined with the earth ellipsoid model(2)This article presents an analysis of the error factors that may affect the accuracy of target positioning.Monte Carlo simulation is used to analyze factors such as line of sight pointing error,geodetic height error of the target,and imaging angle error.In order to address the problem of large errors in positioning in areas with large errors in geodetic elevation based on the Earth ellipsoid model,a digital elevation model is introduced to replace the Earth ellipsoid model.A semi-interpolation algorithm for obtaining the elevation information of the target point is proposed,which improves the efficiency of positioning while ensuring the positioning accuracy in areas with significant terrain fluctuations.(3)In this paper,a real-time on-board positioning and solving module based on DSP is designed.The position information and attitude information of aerial camera are obtained by PPOI-A28 aerial camera positioning and attitude setting system,and the frame Angle information of camera is provided by photoelectric shaft encoder.All the above information is input to DSP to realize real-time target position solving.The applicability of the algorithm is verified by flight test data and remote sensing images.According to the experimental results,the positioning accuracy of the two models based on Earth ellipsoid and digital elevation is roughly the same in the plain with small relief.The accuracy of the algorithms based on digital elevation model are obviously improved in hilly and mountainous areas with large relief.Compared with the traditional DEM algorithm,the proposed elevation half-interpolation iteration method can improve the efficiency of location iteration by 3.9% ～ 45.5% in the area with large target geodetic fluctuation.