Related Concept Videos
Distance Corrections
Improving Translational Accuracy
Improving Translational Accuracy
Proofreading
Proofreading
Proofreading
Errors During Replication are Corrected by the DNA Polymerase Enzyme
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Progressive refinement of 3-d images using coded binary trees: algorithms and architecture.
Recursive implementation of a two-step nonparametric decision rule.
Boundary detection in multidimensions.
Experiments in Text Recognition with Binary n-Gram and Viterbi Algorithms.
An integrated algorithm for text recognition: comparison with a cascaded algorithm.
Variable duration hidden Markov model and morphological segmentation for handwritten word recognition.
Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.
RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.
CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.
DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.
Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.
Learning Shape Anchors for Holistic Indoor Scene Understanding.
Related Experiment Video
Updated: May 23, 2026

Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment
Published on: March 11, 2021
A string correction algorithm for cursive script recognition.
1Department of Computer Science, State University of New York at Buffalo, Amherst, NY 14226.
This study introduces a novel dynamic programming method for correcting noisy text from cursive script recognition. The algorithm improves accuracy by using a trie structure and stack decoding for efficient string estimation.
Area of Science:
- Computer Science
- Artificial Intelligence
- Natural Language Processing
Background:
- Cursive script recognition systems often produce noisy output strings.
- Accurate estimation of the original text from corrupted data is a significant challenge.
Purpose of the Study:
- To develop an effective method for estimating the correct string (X) from a noisy version (Y) generated by cursive script recognition.
- To introduce an accurate channel model accounting for symbol splitting, merging, and substitution.
Main Methods:
- Utilized dynamic programming search combined with stack decoding and a trie structure representation of a dictionary.
- Derived and analyzed the computational complexity of the proposed algorithm.
- Compared the algorithm's performance against methods based on the generalized Levenshtein metric.
Main Results:
- The proposed algorithm demonstrated effective estimation of correct strings from noisy cursive script recognition output.
- Experimental results on English text using a dictionary of common words validated the algorithm's performance.
Conclusions:
- The dynamic programming approach with a trie structure and stack decoding offers an efficient and accurate solution for correcting noisy text in cursive script recognition.
- The developed method provides a valuable tool for improving the reliability of character recognition systems.
