Phase compensation algorithm based on image segmentation in dual-wavelength holographic microscopy.

Abstract

In order to solve the problem of phase compensation errors in the traditional 2π phase compensation method caused by a rough surface and complex structure of objects in dual-wavelength digital holographic microscopy, a phase compensation algorithm based on image segmentation was proposed. First, the phase less than zero in the phase obtained by an equivalent wavelength is compensated for by adding 2π initially. Then the phase after the initial compensation is binarized, and the small connected areas in the binarized graph are removed, so as to obtain a new binarized graph. Finally, according to the two binarized graphs, the phase of the object after the initial 2π phase compensation is recompensated for in different regions, so as to obtain the continuous phase distribution of the object. Based on the dual-wavelength digital holographic microscopy experimental system with an adjustable equivalent wavelength, the proposed algorithm is used to perform three-dimensional imaging of the channel of the microfluidic chip. The experimental results show that the proposed method can effectively obtain the continuous real phase of the object when the structure of the object is known, so as to obtain a more accurate and reliable three-dimensional topography of the object. The above results provide a new idea for the high-quality three-dimensional imaging of the microfluidic system.