Document Image Binarization Using Independent Component Analysis For OCR

The Image binarization plays a vital role in text segmentation which is used in OCR application. Binarization of text in degraded images is a challenging task due to the variations in size, color and font of the text and the results be often affected by complex backgrounds, dissimilar lighting conditions, reflections and shadow. A robust solution to this problem can significantly enhance the precision of scene text recognition algorithms leading to a variety of applications such as scene understanding, navigation, automatic localization and image retrieval. In this paper, we propose a novel method to extract and binarize text as of images that contains complex background. We apply an Independent Component Analysis (ICA) based technique to map out the text region, which is uniform in nature, while removing specularity, shadows and reflections, which are included in the background. This algorithm works better on images with different degradations. We implement our method on various DIBCO datasets.