Support-domain constrained phase retrieval algorithms in terahertz in-line digital holography reconstruction of a nonisolated amplitude object

Phase retrieval algorithms applied to in-line digital holography reconstruction can weaken interference from the region outside the study target and an unstable light source, etc., by adopting the object-plane support domain constraint. Based on threshold segmentation and morphological filtering, a method to directly calculate the object-plane support domain is proposed in this paper. Combined with the above method, an improved support-domain constrained phase retrieval algorithm is presented. Then, imaging simulations and experiments on terahertz in-line digital holography reconstruction of nonisolated objects are conducted. The simulations study the influence of transmittance of the background plate, structural element of morphological filtering, etc., on the reconstruction effect of the improved algorithm without noise interference. Simulation and experiment results suggest that good reconstructed images can be obtained by this algorithm when transmittance of the background plate is greater than 0.90. (C) 2016 Optical Society of America