Latest Issue
The 15th Scientific day : Bridging Science, Engineering, Technology & Industry for Cambodia's Growth
Published: July 31,2026Comparison Analysis of Deep Learning and Tree-Based Homogeneous Ensemble Learning Predictive Models on the KHR/USD Daily Official Exchange Rate
Published: July 31,2026Comparing Time Series, Machine Learning, and Deep Learning Models for Paddy Rice Price Forecasting in Battambang Province, Cambodia
Published: July 31,2026The Forecasting for Cambodia’s Rice Production Using Multiple Linear Regression and Tree – Based Ensemble Learning Methods
Published: July 31,2026From Educational Model to Achievements - Classifying Post-Graduate Success with Machine Learning
Published: July 31,2026Analysis on Machine Learning Models for Imbalanced Data Problem in Payment Fraud Detection
Published: July 31,2026A Computational Approach to Labor Market Analytics: Analyzing Skill Trends in Cambodia Using an LLM-Based Pipeline
Published: July 31,2026Recognizing Khmer Handwritten Digits with the Power of Sequential RNNs
Published: July 31,2026KVerifyID: A Hybrid Multimodal Approach for Khmer Online Writer Verification
Published: July 31,2026Improving Recognition Result Using Character Trigram for Khmer OCR
-
1. Department of Computer Science,
Institute of Technology of Cambodia, Russian Ferderation Blvd., P.O. Box 86, Phnom Penh, Cambodia.
Academic Editor:
Received: January 20,2024 / Revised: / Accepted: January 20,2024 / Available online: June 01,2013
The recognition phase of an Optical Character Recognition (OCR) system produces a ranked list of candidate characters, among which the top one is usally taken as recognition result without taking context into account. Recognition error occurs if the correct character is not at the top, which is mostly due to shape similarity between characters.In this paper we propose to use character trigram, which means that two previous characters are taken into account when choosing the character from the candidate list as recognition result for Khmer OCR.A text corpus of about 300 Mbytes is used to compute character trigrams. Using these trigrams, we test our approach on about 3000 characters. The result shows that this approach can correct about 30% of recognition errors.