An on-line procedure for recognition of handprinted alphanumeric characters

Area of Science:

• Computer Science
• Artificial Intelligence
• Pattern Recognition

Background:

• Handwritten character recognition is crucial for digitizing information.
• Existing methods face challenges with variations in writing styles.

Purpose of the Study:

• To develop and evaluate a robust procedure for recognizing handprinted alphanumeric characters.
• To assess the effectiveness of a two-step classification approach combining structural and statistical methods.

Main Methods:

• Preprocessing and polygon segmentation of input characters.
• Feature vector extraction describing polygon segments.
• Two-step classification: structural analysis followed by statistical decision rules with recursive learning.
• Evaluation using created databases and statistical classifiers.

Main Results:

• The developed procedure demonstrates effective recognition of alphanumeric characters.
• The two-step classification approach, incorporating recursive learning, yields significant recognition rates.
• Analysis confirms the validity of feature vector distributions and potential for simplification.

Conclusions:

• The proposed method offers a viable solution for handwritten character recognition.
• The combination of structural and statistical classification is effective.
• Further simplification through principal axis analysis is feasible.