Research On The Photoelectric Detection Technology Of The Low Altitude,Low Speed,and Small Target

Photoelectric detection technology is a technique for target detection based on the difference between the wavelength and intensity of light waves emitted by target and background.It can detect optical signals from a variety of bands,ranging from ultraviolet(0.2 ~ 0.4?m),visible light(0.4 ~ 0.7?m),infrared light(1 ~ 3?m,3 ~ 5?m,8 ~ 12?m).In this paper a new detection algorithm is proposed according to analyze the infrared characteristics of the low,small and slow target.Because of the low,small and slow target is cheap and easy to get,it often takes off without permission.In recent years,with the increase of the threat situation of the low,small and low target,it needs to be supervised.These targets' species,production materials are diverse,and they are same with background in IR image because they don't have a strong heat source.The commonly used infrared dim target detection algorithm cannot suppress the background adequately.At present,the detection of low small and slow targets is mainly based on radar detection,and there are few infrared detection algorithms.Many research institutions at home and abroad have been carrying out research on infrared low,small and slow target detection.The research contents and achievements of this paper are mainly four points.(1)In this paper,the infrared imaging characteristics of low small and slow targets are analyzed with uav as an example.By analyzing the differences in infrared characteristics between the low,small and slow targets and traditional infrared dim targets,the paper concludes that the low,small,and slow targets have lower SNR and complex trajectories in infrared images.(2)The enhancement of the low,small and slow target was studied.First,the singular value decomposition of the image is performed.Then design a nonlinear function to transform the Singular value matrix.Finally,the contrast of the high frequency part in the reconstructed image is enhanced.In this way,the distinction between goals and backgrounds is more pronounced.(3)To solve the problem that Robinson Guard filter is sensitive to extreme value,the original calculation of Robinson Guard filter was improved.The improved Robinson Guard filter can be more effective to distinguish the foreground and background,for background suppression is more fully.(4)On the basis of the above research,this paper proposes a new infrared target detection algorithm to detect the low,small and slow targets.First,use the enhance method above to enhance the target,and then use the improved Robinson Guard filter for background suppression,finally using adaptive threshold segmentation method based on local contrast(LC)to extract the targets.The real infrared images are used for the simulation experiment,and the results show that the proposed algorithm can obtain a better performance for the target detection.