Open AccessINPUTING STUDENTS’ SCORE BASED ON GIST FEATURES, SUPPORT VECTOR MACHINES AND TESSERACT
Author(s) -
Hau Hung Nguyen
Publication year - 2020
Publication title -
khoa học
Language(s) - English
Resource type - Journals
ISSN - 1859-4816
DOI - 10.35382/18594816.1.41.2020.646
Subject(s) - support vector machine , handwriting , gist , computer science , artificial intelligence , code (set theory) , natural language processing , pattern recognition (psychology) , biology , programming language , set (abstract data type) , stromal cell , cancer research
Handwriting recogination plays an important role in data inputing and processing in the practice. This attracts much attention of many researchers in different fields. In this paper, a new algorithm is proposed by basing on GIST features, Support Vector Machines (SVM) and Tesseract for entering the score on students’ transcript form at Soc Trang Vocational College. The algorithm consists of two main works, i.e., recognizing students’code and recogziing handwritten digit. In the proposed algorithm, all regions of interest are determined and extract their dictint features with using tesseract and GIST. Then, these features are classified by SVM mechanism. Experimental results demonstrated that the proposed algorithm obtained high performance with accuracy up to 96,57% for students’ code and 93,55% for Handwritting scores. Average time was 7,9s per one transcript.