Research On Digital Correction Technology In Optical Joint Design

With the rapid development of electronic technology and material technology, digital imaging system has become the mainstream.The joint application of optical system design and digital enhancement technology provides the concept of combination design, and the concept of simple optical system is proposed by the research of digital enhancement technology.The combination design method has opened up the channel between the traditional optical design and the digital enhancement technology, and provided one kind of closed loop design idea, which had more targeted compared to the existing optical design technology to have.By a computer algorithm to achieve to the point spread function (PSF) which is the key parameter measurement, it can feedback the reliable system performance parameter to the optical module design system, and more by the PSF got,it is convenient and effective to get pictures of debluring processing, saving precious space load, improving the performance of the whole system.Whether for got the image of the super-solution, or removed the blur of the image, the problem is derived from the image frequency aliasing, low quality image overlay, and imaging equipment inherent inherent imaging defects.The PSF is a function that contains all of the imaging mathematical function of fuzzy information.In accordance with the point of view of computer graphics, as the core of the integral,it will be obtained through parameter estimation and machine learning method for approximation.In accordance with the basic classification, it can be divided into parametric and non parametric, blind and non blind distinction for classification.In this paper, the PSF measurement method needs to choose a model without a priori information, which is not regular, and can be used to deal with the problem of frequency aliasing.Because of the need to face the geometric correction, non-uniform lighting and other errors produced so that the final decision to choose to use a certain noise distribution, and can be estimated as much as possible.The most basic method of this choice is to transform a deterioration problem into a good one. This correction mode determines that the quasi optimal white noise model can meet the above objectives.