Research On A Self-collimating And Focusing Technique Based On Image Processing

Aerial camera has the characteristics of high resolution,real time and strong maneuverability.It is widely used in resource exploration,topographic mapping,urban construction and other fields.The aviation environment is complex and changeable.In the process of imaging,the camera is affected by environmental temperature,pressure and imaging distance changes,so the photosensitive surface of the detector deviates from the position of the focal surface of the image,resulting in the decline of the imaging clarity.With the development of modern imaging technology,the focal length of aerial camera gradually becomes longer,the size of detector pixel gradually decreases,and the camera is more and more sensitive to the changes of the environment.Therefore,it is an important research topic in the field of aviation remote sensing to study the influence of the change of aviation environment on the image quality of camera and to propose the high precision and fast automatic detection and focusing method.This paper analyzes and discusses the theoretical basis and development status of automatic focus detection and adjustment technology for aerial cameras.The effects of different focal evaluation functions and target shape and size parameters on the focal accuracy are systematically studied.In order to improve the precision and rapidity of the automatic focusing system,an improved square gradient function operator is proposed.The radial target with 5 degree central Angle was obtained by optimization.Finally,the effectiveness of this method is verified by automatic focusing experiment.The main research contents of this paper are summarized as follows:Firstly,this paper introduces the research background and significance of the automatic focus detection technology of aerial camera.The research status of automatic focus detection method is summarized.Defocusing principle of aerial camera is studied from geometry and wave optics,and defocusing factors of aerial camera are analyzed.The performance characteristics of program control,photoelectric autocollimation and image processing are summarized.A self-collimating method based on image processing is proposed.The focal principle of this method is introduced.Based on the principle of autocollimation,the eigenvalue of the autocollimation target is calculated by using the image focusing algorithm.Secondly,the focus detection characteristics of evaluation function operator are introduced.Simulation analysis was carried out on four types of definition evaluation function operators,namely spatial class,frequency class,informatics,and statistics.The calculation time,steepness and unbiasedness of different evaluation function operators are compared.The sq-grad focusing operator based on spatial domain is improved and used as the focusing algorithm of the focusing system in this paper.Then the space frequency sensitive to defocus of optical system is studied,and the size parameters of radial and grating targets are optimized.The definition evaluation function is used to calculate the evaluation curve of the fuzzy sequence diagram of the target.Finally,the experimental platform of automatic focal detection and external imaging is built,and the improved Sq-Grad focal detection algorithm is verified to have higher focal detection accuracy by external imaging.By using the auto-focus experiment,it is verified that the auto-collimation method based on image processing has a higher precision of 26% and a less time of 32% than the traditional photoelectric auto-collimation method.