Theoretical And Experimental Study On Digital Holography And Holographic Auto-focus Reconstruction

In digital holography, digital hologram is recorded by digital image sensor (such as CCD) instead of photosensitive materials and then saved in computer while intensity and phase reconstruction of the recorded object are performed numerically by computer. Compared with conventional methods, digital holography has some obvious advantages for its low production cost, fast imaging and flexible recording and reconstruction process. For recent years, with the development of computer, especially the development of high resolving digital image sensor CCD, digital holograph with its researches attracts many researchers'attention and has been widely applied in various fields of shape measurement, deformation measurement, digital holomicroscopy etc. Based on theoretic analysis, computer simulation and the corresponding experiment research, this paper gives some analysis about fundamental issue of digital reconstruction of hologram including the determination of involved physical parameter (such as object distance, reference angle and the pixel size of the image sensor etc), the choice of reconstruction algorithms and the location of exact image plane. By analysis above problems, some important research results are obtained. In this paper, the main innovative ideas are as following:1. The influence of the illumination parameters on the quality of reconstructed images in digital reconstruction of an off-axis Fresnel hologram is analyzed. A new algorithm is proposed for extracting the optimal illuminating orientation according to the spatial spectrum of the hologram. The pixel size of the image sensor can be also calculated by use of the parameters of the object distances measured in recording and reconstructing process at the same time.2. Four auto-focus algorithms of digital imaging process are compared and analyzed. And these four algorithms are gray variance method, gray entropy method, Fourier spectrum function method and discrete wavelet transform method. These algorithms are applied in digital hologram reconstruction to determine the accurate location of reconstructed image plane. The results show that discrete wavelet transform method is superior to other three algorithms on sensitivity and accuracy. Propose an algorithm using the variance of intensity on imaging plane as auto-focus criterion according to the characteristic of intensity distribution on imaging plane of phase object; successfully realize auto-focus of reconstructed image for digital hologram imaging process.3. Three computer simulated algorithms of Fresnel diffraction integral are analyzed in detail based on scalar diffraction theory and fast Fourier transform. and these three algorithms are applied in digital reconstruction program for Fresnel holograms. Theoretical analysis proves that, when the pixel number, pixel size and reconstructed wavelength of the hologram are certain, computer simulated algorithms of Fresnel diffraction integral fall into the sampling algorithm of Point spread function, the sampling algorithm of angular spectrum and direct Fourier transform algorithm according to the difference of the sampled object. By analyzing the local spatial frequence of phase function, a proper reconstructed distance used to determine the applicable range of three algorithms is obtained. And above analysis indicated the following results: If the reconstructed distance is larger than proper reconstructed distance, the sampling algorithm of Point spread function is the best algorithms. If reconstructed distance is less than proper reconstructed distance, the sampling algorithm of angular spectrum is the best algorithms. If reconstructed distance is equal to proper reconstructed distance, all three algorithms are suitable.4. Based on analysis for digital reconstruction of an off-axis Fresnel hologram, a complete digital reconstruction program for off-axis Fresnel holograms is designed using GUI interface in MATLAB.