【摘要】 针对无透镜傅里叶变换全息图数值再现过程的特点 ,提出了一套不同于以往的削弱或消除全息图 0级衍射和孪生像的数字图像处理算法。该算法分为三步。首先 ,对所拍摄的数字全息图通过高通滤波实现衬比度增强预处理 ,以提高全息图的衍射效率并消除再现图像的 0级衍射斑 ;其次 ,对再现图像进行带通及V型滤波 ,以降低背景噪声并使再现视场均匀 ;最后 ,对所得到的再现像进行平滑处理 ,通过小波滤噪和中值滤波进一步提高再现像的信噪比。实验结果表明 ,该方法只需记录一幅数字全息图 ,通过简单的数字图像处理 ,便可明显改善图像质量 ,尤其适用于无透镜傅里叶变换全息图的数值再现 ,并且更加实用化。更多还原

【Abstract】 Based on the characteristics of numerical reconstruction for lensless Fourier transform hologram, a set of digital image processing algorithm is presented to eliminate zero diffraction component and “twin image”. This method is different from the former one and consists of three steps: hologram contrast enhancement, bandpass filtering and V-filtering, and reconstructed image smoothing. The hologram contrast enhancement pretreatment, which can be achieved by highpass filtering, is adopted to improve the diffraction efficiency of the recorded digital hologram and suppress zero diffraction component of the reconstructed image. Using bandpass filtering and V-filtering to the reconstructed image with background noise, the noise can be eliminated and the viewfield is made homogeneous. Furthermore, smooth processing such as wavelet filtering and median filtering can also enhance the signal-noise ratio of the reconstructed image. It is shown that the method need record hologram only once and these three simple processing steps are effective for improving the reconstructed image quality and obtaining the satisfying results. It is especially applicable for the numerical reconstruction of lensless Fourier transform hologram, and more practical.更多还原