Font Size: a A A

Research On Aerial Sampling Method Of UAV For Aerial Light Field

Posted on:2024-05-24Degree:MasterType:Thesis
Country:ChinaCandidate:Y C WangFull Text:PDF
GTID:2542307124471964Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Multi-view sampling technology of UAV aerial photography is a key technology in the process of large-scale scene three-dimensional reconstruction,and it is also an important technical means to obtain complete information of aerial light field.Aerial light field rendering technology attracts much attention because it can record complex light in large-scale scenes,and high-quality large-scale scene rendering effect can be obtained by using unmanned aerial vehicle to sample and reconstruct aerial light field.However,the traditional aerial light field rendering technology can not capture the rich texture characteristics of large-scale scenes,and the obtained viewpoint rendering effect is not ideal,and the previous aerial sampling technology relies on over-sampling to improve the rendering quality of scenes,which greatly restricts the development of aerial light field rendering technology.Therefore,how to obtain large-scale scene information efficiently and achieve high-quality scene rendering is an important research problem of aerial sampling technology of UAV based on aerial light field.In this paper,by expanding the traditional light field rendering technology,a sampling method of aerial light field of UAV is designed,which improves the rendering quality of largescale scenes of aerial light field.According to the Fourier theory,the spectrum structure of the airborne light field is obtained,so as to determine the bandwidth of the airborne light field.According to the bandwidth calculation,the minimum sampling rate of aerial light field is obtained,and the high-quality rendering of new viewpoints can be completed while obtaining the viewpoint information of large-scale scenes efficiently.Combined with the characteristics of large-scale scenes,a new anti-aliasing reconstruction filter is designed to further optimize the sampling method of UAV aerial light field.As one of the new research branches of light field rendering technology,the theoretical system of aerial light field rendering technology is not perfect,and there is still room for improvement in the construction of related models.Therefore,aiming at the above problems,this paper studies the aerial sampling method of UAV for aerial light field.The specific research contents are as follows:(1)Determination of sampling method of aerial light field of UAV.Firstly,the UAV sampling model is combined with the aerial light field sampling model to design the aerial light field sampling model of UAV aerial photography.According to this model,Fourier theory is used to analyze the spectrum of airborne light field and determine the influence of specific factors on sampling or spectrum.Through the analysis of frequency spectrum,it is found that with the increase of pitch angle,the frequency spectrum of airborne light field also expands.The bandwidth is obtained by modulo operation on the spectrum expression,and then the minimum sampling rate is obtained,and a new reconstruction filter is designed.(2)Optimization of airborne light field sampling method for UAV.Through the sampling interval of UAV airborne camera determined by the minimum sampling rate,the flight speed and camera pitch angle of UAV sampling are determined.According to the designed sampling method,the UAV is used to sample,and the collected large-scale scene data is processed.Combined with the method of inter-frame estimation,the appropriate viewpoint images are selected from the sampled video,and a new rendering method is designed for multi-viewpoint reconstruction.Experiments show that high-quality large-scale scene rendering can be achieved by using the optimization method of aerial light field sampling in UAV aerial photography.
Keywords/Search Tags:Aerial light field, UAV, Spectrum analysis, light field sampling, view reconstruction
PDF Full Text Request
Related items