Partial Differential Equations are becoming one of the core tools for low-level image processing. They are espe- cially functional in diffusion processes and variational mod- els. In this paper, we exploit the regional smoothing that occurs in a nonlinear diffusion process and use this to en- hance text in a degraded document image. The proposed smoothing method is robust when applied to either a highly corrupted text document or one with little degradation. The technique was tested on historical docu- ments, carbon copies with highly varying grayscale back- grounds and on synthetic noisy documents. The PDE-based method far outperformed other industry-standard binariza- tion techniques when compared quantitatively and qualita- tively.