Document images are usually degraded in the course of photocopying, faxing, printing, or scanning. Degradation problems seems negligible to human eyes but can be responsible for an abrupt decline in accuracy by the current generation of optical character recognition (OCR) systems. In this paper we present a binarization method based on retinex theory followed by a global threshold. The proposed method has been evaluated and compared to other existing methods. The experimental results showed high quality results in terms of visual criteria and OCR performance for the proposed method even for documents captured in an inadequate illumination condition.