<p>Estimation of physical parameters from the noisy image of a Newton’s ring is an important problem in optics. Different methods are available in the literature to estimate unknown parameters from a given image. Among different estimators, it is well known that the least squares estimators are the most efficient estimators in the sense that the asymptotic variances of the least squares estimators reach the Cramer-Rao lower bound when the noise is Gaussian. But, it is well known that the least squares estimators are not robust. They are very sensitive to the outliers. Even in the presence of few outliers the performances of the least squares estimators affect significantly. In this case some of the robust estimators like least absolute deviation estimators or Huber-M estimators may be used. But they are quite difficult to implement in practice. Moreover, they may not work very well when there is outliers in the data set. In this paper we have proposed weighted least squares estimators which can be implemented quite easily and they perform very similar to the least absolute deviation estimators or Huber-M estimators when there are outliers in the data set. Moreover, they perform very similar to the least squares estimators when there is no outliers in the data set. We have derived the asymptotic properties of the weighted least squares estimators. Extensive simulations have been performed to compare the performances of the different estimators. Implementation of the proposed weighted least squares estimators have been suggested. Finally we have provided one illustrative example to show how the proposed method can be used in practice.</p>