An efficient approach for handling degradation in character recognition

Abstract

IJAIP fosters the exchange and dissemination of applications and case studies in the area of advanced intelligence paradigms among education and research professionals. The thrust of the journal is to publish papers dealing with the design, development, testing, implementation and management of advanced intelligent systems, and to provide guidelines in the development/management of these systems. IJAIP publishes archival articles and assessments of current trends, providing a medium for exchanging scientific research and technological achievements accomplished by the international community.

Deep Convolutional Neural Networks for Degraded Printed Kannada Character Recognition

“…Accuracy of 91.3% and 92% is reportedfor printed and handwritten character samples respectively. Sandhya .N et al [6] have proposed a model wherein degradation in characters are measured in scales called high, medium, low. A new method FDA has been employed to compare recognition accuracy with other approaches namely SVM and FLD.The characters have been rebuilt to improve recognition accuracy.…”