Research On The Optimization Of Autofocus Technology Of Liquid Crystal Lens

Clear and sharp images are crucial to the success of later image processing such as image recognition,image segmentation,and image classification.Autofocus technology is a necessary condition for the system to obtain clear and sharp images and is widely regarded as one of the core technologies of optical imaging systems.The miniaturization and intelligence of the imaging system have become the trend.The features of low power consumption,no mechanical movement,and low price of the liquid crystal lens autofocus system make it possible to further miniaturize the imaging system.The current liquid crystal lens auto-focusing method based on the depth of focus focusing technology mainly has the following three problems.First,the limited power range of the liquid crystal lens limits the focusing range of the liquid crystal lens imaging system.Second,because the liquid crystal lens only modulates unusual light,the liquid crystal lens focusing system is usually used in conjunction with a polarizer to remove the interference of ordinary light on the focusing result and the presence of the polarizer reduces the luminous flux of the imaging system and limits the further miniaturization of liquid crystal lens focusing system.Third,because the clear aperture of the liquid crystal lens is inversely proportional to the zoom speed,the zoom speed of common liquid crystal lenses is slow,and the slow focusing speed of the liquid crystal lens focusing system using depth-of-focus focusing technology cannot be applied to application scenarios that require high focusing speed.In this paper,by adjusting the initial focusing distance of the liquid crystal lens imaging system,the liquid crystal lens imaging system can image the entire scene.By introducing the liquid crystal lens unbiased imaging technology into the focusing system,the polarizer is removed to make the system structure more concise.By studying the depth-of-defocus focusing method based on blur,the focus speed of the liquid crystal lens autofocus system is improved.The thesis first analyzes the research background and significance of the subject,and introduces the development history and current status of liquid crystal lenses and autofocus technology,the structure and performance measurement methods of liquid crystal lenses at the TTI level,and the basic principles of the defocus depth focusing method.Then it introduces the optical structure of the current liquid crystal lens focusing system and the principle of autofocus,and introduces representative imagepreprocessing techniques,focus evaluation operators and feedback search algorithms in detail.Finally,it analyzes the problem of current liquid crystal lens focusing system.Then this paper analyzes the influence of the initial focusing distance on the focusing range of the liquid crystal lens focusing system,introduces the LCD lens unbiased imaging technology and improves the original basis to reduce image noise.The unbiased imaging technology is applied to the LCD lens autofocus system to remove Polarizers making the focusing system more compact.Then it analyzes the influence of unbiased imaging technology on Absolute Tenengrad focus evaluation operators,and improves the focus results by using the multi-image average denoising algorithm.Finally,combined with the small aperture and large depth of field of the liquid crystal lens system,the current feedback search step size is optimized.In the end,this paper introduces the principle of defocus depth acquisition of liquid crystal lens based on ambiguity,and proposes a guided Gaussian interpolation algorithm to repair the sparse nature of the depth map.Then combined with the power characteristics of liquid crystal lens and depth map,a depth focusing method based on power voltage fitting was proposed and its accuracy was analyzed.Finally,the accuracy of the defocus depth-focusing system based on blur is analyzed,and its results are corrected using the depth of focus in combination with its characteristics.