Home | Bahasa | Sitemap | HelpDesk | UG-Pedia | Contact
Search
Direktori UG : A - Z
A
B
C
 
Uzbek-Indonesian Joint International Conference, October 8-9, 2013  
Uzbek-Indonesian Joint International Conference On Economics And Management Toward Nation Character Development. Branch Of Russian Economic University After G.V.Plekhanov In Tashkent, Uzbekistan And ...
 

 
Indonesia Open Source Award 2012  
Indonesia Open Source Award (IOSA) 2012. For Further Information http://iosa.web.id/ ...
 



Home > Archive >
Thesis Defence of Doctoral in Information Technology: Naeli Umniati dan Dolly Indra


Doctoral Program Gunadarma University conducted doctoral thesis defence in Information Technology on Wednesday, Nov 29th 2017 at Auditorium of Gunadarma University, Campus Cikunir for promovenda Naeli Umniati and Dolly Indra .Promovenda Naeli Umniati with the supervising commission: Prof. Dr. rer. Nat. A. Benny Mutiara as a promoter and Dr. Tubagus Maulana and Dr. Suryarini Widodo as co-promoter was given the first opportunity to present his dissertation entitled, Development of Offline Signature Identification Method Based on Global and Local Features Extraction.

Signatures are a security technique for identifying someone, because everyone has different signatures and each signature has unique signature characteristics. The problem in an offline signature is the presence of someone in writing a signature. Therefore, it is necessary to know the characteristics of the signature of a person to be recognized properly.

In this study the identification method of identity of the signature offline based on the extraction of global features and local features. The research dataset was obtained from the collection of 100 sets of signatures with each set consisting of ten signatures. Global features captured are the width-height ratio and signature density ratio, while for local features three features are contrast, correlation and homogeneity which are features of Gray Level Co-occurence Matrix (GLCM). The test is performed on scanned data with a resolution of 300 dpi.

The best identification accuracy level was obtained by using global and local features in 300 sets of test data is 77%. Of the 100 signature owners, all test data from 51 people were correctly identified.

The result of the senate session stated that the promovenda Naeli Umniati graduated with a very satisfying predicate.
After a short break, the Gunadarma Senate's hearing continued with the 2nd session session with the promovenda Dolly Indra with the title of dissertation Development of Bisindo Bison Signaling Method Based on Shape Feature Similarity, which was promoted by Prof.Dr. Sarifuddin Madenda and Dr. Eri Prasetyo Wibowo as co-promoter.

The method used in this research is the introduction of the Bisindo letters based on the hand form feature which implies every form of Bisindo letters, which is divided into two parts: the first is the formation of the database form of Bisindo dai A to Z and the second is the introduction of the letters Bisindo.

In the first part consists of segmentation process, edge detection process, feature extraction process in the form of probability value of emergence of hand form chain code and formation of feature database. The second process comprises the process of hand-form image acquisition as a Bisindo letter query, followed by a segmentation process, edge detection process, hand feature extraction process and recognition process by calculating the distance gap of the query feature feature against each feature feature in the feature database. The process of image acquisition in the above two sections is done directly through a Webcam connected on a computer device. This method has been implemented into a prototype interface of luna introduction of Bisindo. The experimental results show that all 100% Bisindo letters can be identified with an average of similarity rates above 95%.

Based on the results of the Senate meeting stated that the promovenda Dolly Indra graduated with a very satisfying predicate.