Handwritten Carbon Form Preprocessing Based on Markov Random Field

Abstract

This paper proposes a statistical approach to degraded handwritten form image preprocessing including binarization and form line removal. The degraded image is modeled by a Markov Random Field (MRF) where the prior is learnt from a training set of high quality binarized images, and the probabilistic density is learnt on-the-fly from the gray-level histogram of input image. We also modified the MRF model to implement form line removal. Test results of our approach show excellent performance on the data set of handwritten carbon form images.

The Impact of Ruling Lines on Writer Identification

“…For example, in Cao and Govindaraju's work [13], the patchbased MRF shape modeling was trained on pre-defined shape patterns to recover the deformations of image shapes. The authors reported significant improvements over traditional approaches using a database of handwritten carbon forms.…”

A probabilistic method for keyword retrieval in handwritten document images

“…We use the MRF based document image preprocessing algorithm [3] to binarize the form image and remove the grid lines from the image. Assuming the binarized objective image is x and the grayscale image is y, we solve the maximum a posteriori (MAP) estimation x = argmax x Pr(x|y) using the Markov Random Fields (MRF).…”

A Model-Based Ruling Line Detection Algorithm for Noisy Handwritten Documents

“…Many techniques work well on clean images of good quality, but their performance deteriorates when lines are badly broken due to light printing or scanning, low input resolutions, or significant overlap with handwriting on the page [5].…”