Repository logo
  • English
  • Català
  • Čeština
  • Deutsch
  • Español
  • Français
  • Gàidhlig
  • Latviešu
  • Magyar
  • Nederlands
  • Polski
  • Português
  • Português do Brasil
  • Suomi
  • Svenska
  • Türkçe
  • Қазақ
  • বাংলা
  • हिंदी
  • Ελληνικά
  • Yкраї́нська
  • Log In
    New user? Click here to register.Have you forgotten your password?
Repository logo
  • Communities & Collections
  • All of DSpace
  • English
  • Català
  • Čeština
  • Deutsch
  • Español
  • Français
  • Gàidhlig
  • Latviešu
  • Magyar
  • Nederlands
  • Polski
  • Português
  • Português do Brasil
  • Suomi
  • Svenska
  • Türkçe
  • Қазақ
  • বাংলা
  • हिंदी
  • Ελληνικά
  • Yкраї́нська
  • Log In
    New user? Click here to register.Have you forgotten your password?
  1. Home
  2. Browse by Author

Browsing by Author "Olorunmaiye, E"

Now showing 1 - 1 of 1
Results Per Page
Sort Options
  • Loading...
    Thumbnail Image
    Item
    Offline yoruba word recognition system based on capsule neural network
    (Ilorin Journal of Computer Science and Information Technology, 2022-06-30) Ajao, J. F.; Olorunmaiye, E; Mabayoje, M. A.; Yusuff, S. R.; Bajeh, A.
    The need for Digital Document archiving has resulted in several efforts in Image Document Analysis, where scanned historical/ indigenous documents can be stored in digitized form and made readily accessible and available for future use. Visual system in human uses an highly space variant in sampling, coding, processing and understanding of variations in human handwritings, which possess lots of challenges or difficulty for machine to understand. Consequently, it is important to preserve our indigenous heritage values by way of digitization. In this regard, the research work models a Yoruba Script Recognition System using Capsnet, based on the peculiar features of Yoruba alphabet. Yoruba bird names were gathered from the Yoruba dictionary and were written by 20 indigenous literate writers in 50 different writing styles. This was used to model HYSR system. The training time and testing after training different epoch produces 89.2222 and the testing time generated is 0.8667. The model accuracy gives 85.49%, which shows a promising result.

KWASU Library Services © 2023, All Right Reserved

  • Cookie settings
  • Send Feedback
  • with ❤ from dspace.ng