ARCHIVES
VOL. 10, ISSUE 4 (2025)
Development of an enhanced hyper heuristic Firefly Algorithm based Convolutional Neural network for handwritten identification system
Authors
Locksley EA, Olabiyisi SO, Ganiyu RA, Omidiora EO, Taiwo CY
Abstract
Handwritten recognition is very important
especially in the area of computer vision and pattern recognition and forensic
analysis where the identification of the author of a document is required for
legal/tactical processing. The aim of this work is to enhance Hyper Heuristic
Firefly Algorithm for any handwritten document to enhance the recognition
accuracy. The Zebra Optimization Algorithm (ZOA) was then improved the HHFA for
better optimization, and more accurate results in this study. Hand-written responses
were obtained from employees and students of Ladoke Akintola University of
Technology, Ogbomoso, Nigeria. CNN hyper parameter was optimized using the
improved hyper heuristic firefly algorithm (HHZOFA). The simulated model was
constructed in a MATLAB 2020. The performance of the output was compared to a
hyper heuristic firefly algorithm in existence, and a hypothesis was made based
on t test at p <0.5. Then we had some interesting discovery that the
improved HHZOFA is of high precision and there is significant difference among
HHZOFA and the traditional HHFA
Download
Pages:18-26
How to cite this article:
Locksley EA, Olabiyisi SO, Ganiyu RA, Omidiora EO, Taiwo CY "Development of an enhanced hyper heuristic Firefly Algorithm based Convolutional Neural network for handwritten identification system". International Journal of Advanced Scientific Research, Vol 10, Issue 4, 2025, Pages 18-26
Download Author Certificate
Please enter the email address corresponding to this article submission to download your certificate.