EKSTRAKSI CIRI TEKS PADA GAMBAR ALAMI ENGGUNAKAN ALGORITMA RANDOM FERNS DAN CONVOLUTIONAL CO-OCCURRENCE HOG

DANIEL, MUHAMMAD and Samsuryadi, Samsuryadi and Primanita, Anggina (2018) EKSTRAKSI CIRI TEKS PADA GAMBAR ALAMI ENGGUNAKAN ALGORITMA RANDOM FERNS DAN CONVOLUTIONAL CO-OCCURRENCE HOG. Undergraduate thesis, Sriwijaya University.

Preview	Text RAMA_55201_09111002018_ 0004027101_ 0206088901_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (2MB) | Preview
	Text RAMA_55201_09111002018_ 0004027101_ 0206088901_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (518kB) | Request a copy
	Text RAMA_55201_09111002018_ 0004027101_ 0206088901_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09111002018_ 0004027101_ 0206088901_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (695kB) | Request a copy
	Text RAMA_55201_09111002018_ 0004027101_ 0206088901_05_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (93kB) | Request a copy
	Text RAMA_55201_09111002018_ 0004027101_ 0206088901_06_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (176kB) | Request a copy

Abstract

A natural image is images that represents what the human eye sees daily. The recognition of text in natural images has many obstacles, such as lighting, shooting angle, and variations of fonts used. The process of recognizing a word in a natural image consists of three stages, detecting the location of characters in the image, recognizing detected characters, and combining recognized characters into words. This study is focused in feature extraction process of character detection using Random Ferns method, and recognition of detected characters using the Convolutional Co-Occurrence Histogram of Oriented Gradient method, which was trained and tested on the ICDAR dataset 2003. The results of this study provide a high accuracy on detected words.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Feature Extraction, Text Recognition, Natural Image, Random Ferns, Convolutional Co-Occurrence Histogram of Oriented Gradient.
Subjects:	R Medicine > R Medicine (General) > R858-859.7 Computer applications to medicine. Medical informatics
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Mrs Sri Astuti
Date Deposited:	12 Sep 2019 07:31
Last Modified:	12 Sep 2019 07:31
URI:	http://repository.unsri.ac.id/id/eprint/7247

Actions (login required)

View Item