Tertia, Devara and Nugroho, Anto S. and Galinium, Maulahikmah (2017) Text Localization And Recognition Of Identification Card. Bachelor thesis, Swiss German University.

Preview	Text Devara Tertia 11201001 TOC.pdf Download (394kB) | Preview
	Text Devara Tertia 11201001 1.pdf Restricted to Registered users only Download (322kB)
	Text Devara Tertia 11201001 2.pdf Restricted to Registered users only Download (533kB)
	Text Devara Tertia 11201001 3.pdf Restricted to Registered users only Download (738kB)
	Text Devara Tertia 11201001 4.pdf Restricted to Registered users only Download (1MB)
	Text Devara Tertia 11201001 5.pdf Restricted to Registered users only Download (199kB)
Preview	Text Devara Tertia 11201001 Ref.pdf Download (94kB) | Preview

Abstract

Identification card has been the main reference for officials to identify a person. To be able to make a computer read the card optically, we can automatically reduce effort to read it manually. Therefore one way to achieve this is to utilize image processing techniques and classification techniques to make a computer to read identification card automatically. Several techniques such as Hough Transform, used to accomplish the goal to read identification card automatically. These techniques has proven their versatility in many other research and widely implemented and gives good result. Hough Transform used to detect all possible lines in an image, this can help us to detect card in the image by analyzing all the lines detected. Hough Transform takes an input of an edge image, therefore we use Canny edge detector. Canny edge detector is an enhancement of Sobel operator with the image being smoothed first using Gaussian smoothing. Since each characters extracted first individually, each characters then recognized using Nearest Neighbor algorithm and the result is compared to Naïve Bayes algorithm. The result give good result on detecting the cards as all cards tested successfully being extracted from the image acquired. Although not all the characters are detected some characters extracted recognized incorrectly, therefore this should be able to be edited by human to correct the mistakes. By using these techniques, it is expected to achieve this goal to be able to help working on identification card’s information to ease and make administrative processes faster.

Item Type:	Thesis (Bachelor)
Uncontrolled Keywords:	Text Localization; Image Processing; Canny Edge Detector; Hough Transform; Gaussian
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing
Divisions:	Faculty of Engineering and Information Technology > Department of Information Technology
Depositing User:	Astuti Kusumaningrum
Date Deposited:	11 May 2020 04:34
Last Modified:	11 May 2020 04:34
URI:	http://repository.sgu.ac.id/id/eprint/255

Actions (login required)

View Item