Adaptive–interpolative binarization with stroke preservation for restoration of faint characters in degraded documents

• A novel technique of binarization with stroke preservation for faint characters.
• An adaptive–interpolative thresholding technique in a multi-scale framework.
• Exhaustive experimentation has been carried out with benchmark datasets.
• The proposed method is found to be robust and better than other existing methods.

A novel technique for binarization with stroke preservation of faint characters in degraded documents is proposed. It works in a multi-scale framework with an adaptive–interpolative thresholding technique. Instead of computing a global threshold value, it computes the local threshold values for a small set of grid points by observing the intensity pattern of the pixels lying in the concerned grid cells. Estimated thresholds are used, in turn, to compute the threshold values of all the remaining pixels using a fast-yet-efficient interpolation procedure. To handle noises in degraded images, this grid-based adaptive thresholding is applied in successively reducing scales to obtain the near-optimal binarization as a set of connected components. After a post-processing meant for stroke preservation with these connected components, we get the final output. Exhaustive experimentation and comparison with other existing methods have been successfully carried out with benchmark datasets and also with our own datasets.