Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Preprocessing of low-quality handwritten documents using Markov random fields.

Huaigu Cao1, Venu Govindaraju

  • 1Center for Unified Biometrics and Sensors (SUBS), Department of Computer Science and Engineering, University at Buffalo, Amherst, NY 14260, USA. hcao@bbn.com

IEEE Transactions on Pattern Analysis and Machine Intelligence
|May 16, 2009
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Chart Mining: A Survey of Methods for Automated Chart Analysis.

IEEE transactions on pattern analysis and machine intelligence·2020
Same author

Probabilistic Assessment of Glass Forming Ability Rules for Metallic Glasses Aided by Automated Analysis of Phase Diagrams.

Scientific reports·2019
Same author

Labeling Spain with Stanford.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2013
Same author

Offline Arabic handwriting recognition: a survey.

IEEE transactions on pattern analysis and machine intelligence·2006
Same author

Hidden Markov models combining discrete symbols and continuous attributes in handwriting recognition.

IEEE transactions on pattern analysis and machine intelligence·2006
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Task-KV: Task-aware KV Cache Optimization via Semantic Differentiation of Attention Heads.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Achieving Text-based Person Retrieval with Any Granularity.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces a statistical method using Markov Random Fields (MRF) for cleaning degraded handwritten forms. The approach effectively removes binarization noise and preprinted lines, improving image quality for handwritten data.

Area of Science:

  • Computer Vision
  • Image Processing
  • Machine Learning

Background:

  • Degraded handwritten forms pose challenges for accurate data extraction.
  • Existing preprocessing methods often struggle with noise and background elements like ruling lines.

Purpose of the Study:

  • To develop a robust statistical approach for preprocessing degraded handwritten forms.
  • To improve the accuracy of binarization and remove preprinted ruling lines from historical documents.

Main Methods:

  • Modeling degraded images using Markov Random Fields (MRF).
  • Learning prior probabilities from high-quality images and observation densities from input image histograms.
  • Modifying the MRF to specifically remove preprinted ruling lines.
  • Utilizing patch-based topology and Belief Propagation (BP) for efficient processing.

Related Experiment Videos

  • Pruning the solution space to accelerate MRF computation.

    • Main Results:

    • Achieved higher accuracy in binarization and line removal compared to existing methods.
    • Demonstrated effectiveness on two distinct datasets of degraded handwritten images.
    • The modified MRF model successfully separated handwritten content from background ruling lines.

    Conclusions:

    • The proposed statistical MRF-based approach offers a significant improvement for preprocessing degraded handwritten forms.
    • Efficient processing achieved through Belief Propagation and solution space pruning.
    • This method enhances the quality of historical handwritten documents for subsequent analysis.