We propose a computational method for generating sequential kinoforms of real-existing full-color three- dimensional （3D） objects and realizing high-quality 3D imaging. The depth map and color information are obtained using non-contact full-color 3D measurement system based on binocular vision. The obtained full-color 3D data are decomposed into multiple slices with RGB channels. Sequential kinoforms of each channel are calculated and reconstructed using a Fresnel-diffraction-based algorithm called the dynamic- pseudorandom-phase tomographic computer holography （DPP-TCH）. Color dispersion introduced by different wavelengths is well compensated by zero-padding operation in the red and green channels of object slices. Numerical reconstruction results show that the speckle noise and color-dispersion are well suppressed and that high-quality full-color holographic 3D imaging is feasible. The method is useful for improving the 3D image quality in holographic displays with pixelated phase-type spatial light modulators （SLMs）.
M-arrays are random arrays in which an appropriate sub-window appears only once in the whole array. Coded structured light based on M-arrays is one=shot technique to rapidly acquire 3D information of unknown surfaces by projecting suitable patterns onto a measuring surface. This paper presents a method to construct large size M-arrays based on the piece growing algorithm in which an array is constructed by many pieces through splicing each other. Reconstructing 3D shapes by utilizing the designed pattern based on constructed M-arrays for two objects are given.
