Off-Line Handwritten Bilingual Name Recognition for Student Identification in an Automated Assessment System

Publication Type:
Conference Proceeding
Citation:
Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR, 2014, 2014-December pp. 271 - 276
Issue Date:
2014-01-01
Full metadata record
Files in This Item:
Filename Description Size
off.pdfPublished version314.63 kB
Adobe PDF
© 2014 IEEE. The Student name Identification System (SIS) proposed here was investigated for English and Thai languages combined. The proposed system recognises each name by using an approach for whole word recognition. In the proposed system, the Gaussian Grid Feature (GGF), and Modified Direction Feature (MDF), together with a proposed hybrid feature extraction technique called Water Reservoir, Loop and Gaussian Grid Feature (WRLGGF) were investigated on full word contour images of each name sample. Artificial neural networks and support vector machines were used as classifiers. An encouraging recognition accuracy of 99.25% was achieved employing the proposed technique compared to 98.59% for GGF, and 96.63% using MDF.
Please use this identifier to cite or link to this item: