Research On Optical Microscopic Auto-focus Based On Visual Image

The auto-focus technology is in the machine intelligent,driven by the intelligent automation of the machine.It is widely used in the fields of daily life,scientific research,industrial production and military application involving imaging systems.The Key Technology in Modern Photoelectric System.Since the early 20 th century,domestic and foreign experts and scholars around the issue of autofocus have done a lot of research,making rapid development of the various theories and methods of automatic focusing technology,but because of imaging targets and imaging conditions of the diversity of existing imaging,the current auto-focus technology is not very mature,still need to be improved.How to select the appropriate resolution evaluation function and participate in the calculation of the region,how to balance the accuracy and speed of focus,there are still many problems to be solved.In this thesis,the key technology of microscopic autofocus based on visual image is studied to improve the efficiency and performance of microscope AF.In this thesis,we first compare and analyze the image clarity evaluation function.In view of the above-mentioned problems,a fast autofocus method combining image method and contrast method is adopted,and the parameters in the experimental scheme are simulated.Finally,The feasibility of the method is verified.The main work of this dissertation includes the following aspects:(1)Summarizing a plenty of work of the domestic and foreign experts and scholars,introduced the development of auto-focus technology and research status quo.The structure model and basic principle of optical focusing are described.The principle of microscope focus autofocus based on contrast detection is described,and the corresponding mathematical model is established.(2)The choice of the clarity evaluation function in the AF system directly affects the accuracy and computational efficiency of the AF.In this paper,the method of digital image preprocessing is discussed,mean and median filtering are compared by experiments.Then,several typical image clarification evaluation algorithms are analyzed,and some typical algorithms,Unbiased,unbiased,sensible and computational quantities are compared and analyzed,and the difference of the operators with different resolution is proposed and the optimal algorithm is found.(3)The defects of microscope autofocus technology which is be based on contrast detection processing is proposed,and improve the microscope autofocus method basedon the visual image,by adding optical path difference prism in the optical path,and the use of projected fringe method,access the real-time defocusing direction and the defocus amount.While taking advantage of Zemax optical simulation software simulation parameters used by the system to optimize the design of the optical system to get the best experimental parameters.(4)The structure of the autofocus optical path of the microscope is designed and constructed and experimental research is carried out,verifying the correctness of the program.According to the simulation results of the system device selection,build a microscope AF device by optical microscopy and image sensors,and development of the corresponding measurement software.The feasibility of the proposed method is validated by experiments,and the corresponding search methods are also provided.