Research On Uav Multi-spectral Remote Sensing System Based On Correlation Imaging

With the continuous development of UAV and micro-miniature spectral sensor technology,UAV remote sensing,as a new type of remote sensing platform,has the advantages of low cost,good real-time performance,simple operation,small impact on the weather environment,and high spatial and temporal resolution.Effectively make up for the lack of satellite aerial remote sensing.UAV equipped with spectral imaging equipment is very suitable for real-time acquisition of remote sensing data in a small area,and can be used in agricultural monitoring,land surveying and mapping,environmental monitoring,traffic monitoring and other fields,and has become a research hotspot in recent years.This paper designs and develops a UAV multi-spectral remote sensing system based on correlation imaging technology.It uses a six-rotor UAV platform and uses a multispectral correlation imaging principle prototype to replace the traditional multi-spectral camera in the UAV multi-spectral remote sensing system.The number of spectrum bands of the system has been increased from the traditional 4-6 to 15,achieving single-shot to obtain a three-dimensional spectrum image of the target object,and completing the corresponding field experimental test research work.In view of the limited number of bands and wider bands of current traditional multi-spectral imaging sensors,this system satisfies the need for precise quantitative measurement in UAV remote sensing.The main research contents of this paper are as follows:1.The research on multi-spectral correlation imaging with laser active illumination has been carried out.The experimental results show that compared with the multi-spectral reconstruction image under low illumination condition,the reconstruction image quality of the corresponding laser wavelength spectrum segment has improved obviously when the target object is irradiated by the laser source with different wavelengths.2.The research on multi-spectral correlation imaging of moving targets has been carried out.The experimental results show that as the single-frame CCD detection signal increases with the mean value of electrons,the relative root-mean-square error r RMSE value of the reconstructed image is correspondingly reduced,and the quality of the reconstructed image is improved;When the motor drives the target object to move continuously at a speed of 30 Hz,a multi-spectral reconstructed image of the moving object with better quality can be obtained.3.Based on the principal component analysis method to carry out the numerical simulation research of the correlation imaging,the simulation results show that: through the principal component analysis of the measurement matrix,the original target image is reconstructed by the second-order correlation algorithm,and better image reconstruction results are obtained;For color target objects,an initial measurement matrix with a size of 2048×2048 is used,and high-quality reconstructed image results can be obtained with2500 samples.4.Designed a set of UAV multi-spectral remote sensing system based on correlated imaging(including air system,link system and ground system),and completed the development of related software and hardware systems.The UAV multi-spectral remote sensing system is easy to operate,has a large number of bands,and has a fast imaging speed.It can acquire multi-spectral images with single-shot,and can achieve fixed-point data collection according to the actual requirements of remote sensing operations.5.Completed the field experimental study of the UAV multi-spectral remote sensing system based on correlation imaging.The results show that the two experimental reconstruction results of ground staring shooting and UAV mounted hover shooting are basically the same,verifying that the main performance indicators of the system have reached Design plan requirements.The feasibility of the system is verified based on the indoor static principle verification experiment and outdoor experiment.