Comparison Between Neural Network and Support Vector Machine in Optical Character Recognition | Zendy

Michael Reynaldo Phangtriastu | Zendy; Jeklin Harefa | Zendy; Dian Felita Tanoto | Zendy

AI Assistant Blog Pricing

Home ZAIA Blog

Open Access

Comparison Between Neural Network and Support Vector Machine in Optical Character Recognition

Author(s) -

Michael Reynaldo Phangtriastu,

Jeklin Harefa,

Dian Felita Tanoto

Publication year - 2017

Publication title -

procedia computer science

Language(s) - English

Resource type - Journals

SCImago Journal Rank - 0.334

H-Index - 76

ISSN - 1877-0509

DOI - 10.1016/j.procs.2017.10.061

Subject(s) - support vector machine , computer science , artificial intelligence , pattern recognition (psychology) , artificial neural network , projection (relational algebra) , zoning , histogram , histogram of oriented gradients , feature vector , classifier (uml) , feature extraction , feature (linguistics) , computer vision , image (mathematics) , algorithm , linguistics , philosophy , political science , law

Optical Character Recognition is one of the popular area in artificial intelligence and pattern recognition area. Generally, this technique converts the input image into an editable format in computer. This paper uses several techniques as a comparison for some extracted features, such as: zoning algorithm, projection profile, Histogram of Oriented Gradients (HOG) and combination of those feature extractions ( zoning + projection, projection + HOG, zoning + HOG, zoning + projection + HOG). For the evaluation of the proposed system, this paper compare the most commonly classifiers: Support Vector Machine (SVM) and Artificial Neural Network (ANN). This experiment achieves the highest accuracy of 94.43% using Support Vector Machine (SVM) classifier with the feature extraction algorithms are:projection profile and the combination of zoning + projection profile

The content you want is available to Zendy users.

Already have an account? Click here to sign in.

Having issues? You can contact us here

Accelerating Research