Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Discriminative learning quadratic discriminant function for handwriting recognition.

Cheng-Lin Liu¹, Hiroshi Sako, Hiromichi Fujisawa

¹Central Research Laboratory, Hitachi, Ltd. Tokyo 185-8601, Japan. liucl@crl.hitachi.co.jp

IEEE Transactions on Neural Networks

|September 24, 2004

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

Molecular insight into the interplay among heterogeneous plasmacytes and microenvironment cells and their clinical relevance in myeloma.

Proceedings of the National Academy of Sciences of the United States of America·2026

Same author

Local hyperthermia for cutaneous sporotrichosis: A randomized clinical trial.

Journal of the American Academy of Dermatology·2026

Same author

Low-temperature and rapid determination of water content in solid samples using a MAPbBr<sub>3</sub>/SiO<sub>2</sub> paper-based sensor.

Mikrochimica acta·2026

Same author

Privacy-Preserving Average-Tracking Control for Multi-Agent Systems with Constant Reference Signals.

Entropy (Basel, Switzerland)·2026

Same author

A primal-dual approach to double-risk-constrained LQR for practical control under non-Gaussian noise.

ISA transactions·2025

Same author

From System 1 to System 2: A Survey of Reasoning Large Language Models.

IEEE transactions on pattern analysis and machine intelligence·2025

Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013

Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011

Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011

Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011

Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011

Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011

See all related articles

This study introduces a new discriminative learning quadratic discriminant function (DLQDF) for character string recognition. DLQDF enhances classification accuracy while maintaining strong resistance to noncharacters, outperforming previous methods.

Area of Science:

Computer Science
Machine Learning
Pattern Recognition

Background:

Character string recognition requires high classification accuracy and noncharacter resistance.
Modified Quadratic Discriminant Function (MQDF) offers good noncharacter resistance but lower accuracy than neural networks.

Purpose of the Study:

To develop a discriminative learning algorithm to optimize MQDF parameters.
To improve classification accuracy while preserving noncharacter resistance.

Main Methods:

Proposed a discriminative learning algorithm for MQDF, termed DLQDF.
Optimized DLQDF parameters using the minimum classification error (MCE) criterion.
Assumed Gaussian density for DLQDF parameters.

Main Results:

Related Experiment Videos

DLQDF demonstrated comparable or superior performance to neural classifiers in handwritten digit and numeral string recognition.
The proposed DLQDF achieved competitive results against state-of-the-art methods.
DLQDF successfully balanced classification accuracy and noncharacter resistance.

Conclusions:

The discriminative learning approach effectively enhances MQDF for character recognition tasks.
DLQDF presents a promising alternative to neural networks, offering a balance of accuracy and robustness.
The method is validated for practical applications in handwritten character recognition.