Research Of Auto-focusing Technology In Optical Holography Scanning System

Optical holography is a novel technology for object image recognition. It can be utilized to record three-dimensional(3D) volume information. With the fast development of digital devices and techniques, optical holography has gradually leaded its way to digital optical holography, which is known as digital holography. As an important kind of digital holography technique, optical scanning holography(OSH) can record information of an object into the computer with the help of the digital signal devices. The 3D volume information of the object can be recorded into a 2D hologram via the 2D scanning.As an important field of optical holography, OSH has many advantages such as real-time processing capability, as well as high resolution. It has already become a popular research area around the world. OSH has already found wide applications, such as optical microscopy, image encryption, as well as remote sensing. The thesis is based on the theories in OSH, such as hologram recording, reconstruction and defocus noise suppression, etc, and its main purpose is to research on the auto-focusing technique. The advantages and disadvantages of relevant algorithms have been analyzed and new algorithm have been proposed and verified via simulations and experiments.The main research areas of this thesis can be categorized into the following points:(1) The basic theories of OSH have been analyzed based on the Fresnel diffraction;the problems of the traditional reconstruction method have been discussed, and the corresponding defocus noise suppression algorithm has been proposed. The algorithm is based on the random phase pupil and can effectively suppress the defocus noise as well as to improve the signal-to-noise ratio; the whole recording, reconstruction, and defocus noise reconstruction suppression process have been simulated and verified with Matlab.(2) The advantages and disadvantages of the auto-focusing method based on the Wigner distribution have been discussed based on the basic theories of Wigner function and Hough transform. The proposed Wigner-Hough algorithm has been verified with simulation via Matlab.(3) The principles of various edge detection algorithms have been introduced, such as Roberts, Sobel and Wavelet transform, etc. The auto-focusing method based on discrete wavelet transform and connected component auto-focusing algorithm(DWTCC) hasbeen proposed. It has been demonstrated that the evaluation function of the DWTCC algorithm has better effect both on the in-focus image edge extraction and defocus noise suppression based on the comparison of different edge detection methods.