摘要

To extend the depth of field, the all-in-focus image can be achieved via the computational imaging method by designing the data acquisition process under the condition of maintaining the signal-to-noise ratio. In this paper, the data acquisition of all-in-focus sweep imaging is described by the Wigner distribution WDF); then, the WDF-based all-in-focus imaging model and the computational imaging method are established. In the all-in-focus imaging model, the WDF-based light propagation model is utilized to describe the imaging process. As a consequence, the point spread PSF) of the imaging system is derived via the WDF-based light propagation model. By analyzing the PSF of the imaging model, the approximate 3-D spatial invariance of the PSF is achieved, and the all-in-focus sweep data can be expressed as a convolution of the all-in-focus image and the PSF. As a result, the all-in-focus image can be reconstructed by utilizing the deconvolution in the computational imaging. In the simulated experiments, the computational imaging results in different object planes are analyzed based on the structural similarity index to verify the all-in-focus sweep imaging method. In the real data experiments, the resolution targets placed in two different depths can be imaged with extended depth of field via the all-in-focus imaging method proposed in this paper.

全文