| |
| During the calibration of a Stokes polarimeter, the noisy measurement data will degrade the accuracy of the estimated system matrix. In this paper, by taking into account the physical constraints in terms of radiometric calibration and system response, two constrained estimation methods are proposed to improve the calibration accuracy in the presence of signal-independent additive Gaussian noise, signal-dependent Poisson shot noise, or their combination. Both the numerical and experimental results show that the proposed constrained Maximum-likelihood (ML) estimation has the best accuracy, especially when the noisy measurement data have low signal-to-noise ratio, and the proposed constrained Least-Square (CLS) estimation is more effective than the conventional unconstrained Least-Square (ULS) estimation. The proposed estimation methods are useful for the calibration of the Stokes polarimeter with high accuracy. |
|
| Keywords:polarimetry; Stokes parameter; calibration; Gaussian noise; Poission noise |
|
