Handwriting based on feature extraction of hough transform / Nor Hanim Mokhtar

Nor Hanim, Mokhtar (2016) Handwriting based on feature extraction of hough transform / Nor Hanim Mokhtar. Degree thesis, Universiti Teknologi MARA Cawangan Perak.

Abstract

Handwriting is the expressing of language, just like speech and it also leaves a lasting trace. Some people call handwriting as the „Language by the Hand‟. Besides that, to get similar handwriting is so difficult because everyone have different handwriting. Handwriting can be classified such as margin, slant, and size of the handwriting. The aim for this project is to extract the structure of handwriting using HT technique. The FEHT application, HT method has been applied due to the no research reported by applying this technique for Malay and English handwriting feature extraction. Besides that, this technique is used to detect the circle and line on handwriting image. This technique has been implementing on the original and rotate image in order to evaluate its performance. From the study, the result shows that the HT technique is not very suitable to be implemented to extract the structure of handwriting. The future work for this project are applying different method of feature extraction to extract the structure of handwriting, analyze HT technique based on noise removal and extend the methodology until the classification and recognition stage.

Metadata

Item Type: Thesis (Degree)
Creators:
Creators
Email / ID Num.
Nor Hanim, Mokhtar
2013917473
Contributors:
Contribution
Name
Email / ID Num.
Thesis advisor
Normah, Mohd Rawi
UNSPECIFIED
Subjects: A General Works > Academies and learned societies (General)
Divisions: Universiti Teknologi MARA, Perak > Tapah Campus > Faculty of Computer and Mathematical Sciences
Keywords: Handwriting based, feature extraction, hough transform
Date: 29 January 2016
URI: https://ir.uitm.edu.my/id/eprint/32700
Edit Item
Edit Item

Download

[thumbnail of 32700.pdf] Text
32700.pdf

Download (265kB)

Digital Copy

Digital (fulltext) is available at:

Physical Copy

Physical status and holdings:
Item Status:

ID Number

32700

Indexing

Statistic

Statistic details