Enhanced spectral-domain phase microscopy for high-sensitivity and broad-range quantitative phase imaging via joint algorithm-hardware calibration

Spectral-domain phase microscopy (SDPM) enables high-sensitivity quantitative phase imaging by retrieving phase information directly from the interference spectrum, mitigating the axial-resolution penalty caused by spectral leakage in the discrete Fourier transform (DFT). However, conventional SDPM remains highly sensitive to abrupt phase gradients and noise, often leading to wrap mismatches and global bias that restrict its measurement range. In this Letter, we propose enhanced SDPM (eSDPM), an algorithm-hardware co-calibration framework that achieves high precision, bias-free phase reconstruction over an extended measurement range. On the algorithmic side, we explicitly model endpoint terms in the spectral phase, estimate a reference phase within a safe operating band, and determine a wavenumber anchor from the centroid of the source spectrum. On the hardware side, controlled spectral truncation and boundary-wavelength calibration minimize endpoint bias and wavelength misalignment. Experiments on biological and industrial samples demonstrate that eSDPM achieves continuous, bias-free, and high-accuracy optical path difference reconstruction over a broad measurement range, establishing it as a robust and practical approach for wide-range quantitative phase imaging.