Enhancement of old images and documents by digital image processing techniques

Documents can be a valuable source of information but often they suffer degradation problems, especially in the case of historical documents, such as strains, background of big variations and uneven illumination, ink seepage, etc. Binarization techniques should be applied to remove the noise and improve the quality of the documents. Collections of historical and old document images care commonly provided to public through digital libraries. Specialized processing is required to these document images for removing background noise in order to become more legible. A hybrid binarizatin approach is proposed in this paper for improving the quality for the old documents.

Combination of global and local thresholding techniques are used for the same. Initially, a technique named global thresholding is applied to the whole image. The image area that still has background noise are detected and the technique is again re-applied to each area separately. Therefore, a better adaptability is achieved for the algorithm where various kinds of noise re exist in different areas of same image. Advantage of applying global thresholding, is that it avoids the computational and time cost of applying a local thresholding in the entire image. Hence it is indicated that this technique is pretty effective in removing background noise and improving the quality of degraded images.