A Downhill Simplex Algorithm for Estimating Morphological Degradation Model Parameters

Noise models are crucial for designing image restoration algorithms, generating synthetic training data, and predicting algorithm performance. However, to accomplish any of these tasks, an estimate of the degradation model parameters is essential. In this paper we describe a parameter estimation algorithm for a morphological, binary image degradation model. The inputs to the estimation algorithm are i the degraded image, and ii information regarding the font type italic, bold, serif, sans serif. We simulate degraded images and search for the optimal parameter by looking for a parameter value for which the neighborhood pattern distributions in the simulated image and the given degraded image are most similar. The parameter space is searched using the Nelder-Mead downhill simplex algorithm. We use the p-value of the kolmogorov-Smirnov test for the measure of similarity between the two neighborhood pattern distributions. We show results of our algorithm on degraded document images.