On 11th August 2012, Doctoral program Gunadarma University proudly present the 48th and 49th doctoral thesis defence in Information Technology at auditorium of Gunadarma University, Campus D, Margonda Raya Street, Pondok Cina, Depok. As the first candidat, Hustinawaty presented her thesis dissertation with title "The Development of Skeletonization Algorithm and Non-Intrusive Feature Extraction on Human Gait in Real Time". The advisors committee are Prof. Dr. Sarifuddin Madenda as a Promotor and Dr. Tb. Maulana Kusuma, SKom., MEngSc as a Co-Promotor, and as external examiner is Dr. Hermawan.
This research proposes method and algorithm of human gait analysis as well as the development of prototype software. The developed method is in the form of real-time and non-intrusive (without using markers or without operator intervention) gait analysis. This method consists of image acquisition algorithms, segmentation algorithm, skeletonization algorithm, and motion feature extraction algorithm.
Based on the analysis and testing results through experiments, it can be concluded that the acquisition algorithm could sample the image of human gait in real-time with sampling rate of 30 fps (frame-per-second). The segmentation algorithm was able to produce silhouettes that are relatively similar to those recorded in body shape. Skeletonization algorithm was successfully producing skeleton from the silhouette images, which could simplify feature extraction process. The developed feature extraction technique was successfully performed human gait features calculation automatically and relatively fast without operator intervention or called non-intrusive. The level of accuracy of the developed method and algorithm could reach 97.75%
The second candidate is Latifah with Prof. Dr. Djati Kerami as The Promotor, Dr. Ernastuti as CoPromotor, and Dr. Kiki Aryanti Sugeng, as a external examiner. She defended her thesis dissertation with title "Interconnection Network Topology Torus-Butterfly: Evaluation Through The Parameter of Structural, Enumeration, Embedding and Hamiltonian Path"
This research was designed a network topology by using a Cartesian product of two properties of the network topology that has constant degree, that is the Torus topology and Enhanced Butterfly topology.
To investigate whether the network topology was designed to have better properties than the existing network topology, then evaluated and analyzed parameters: enumeration, structural Hamiltonian path, embedding and combinatoric recursive structure of the graph. Of evaluation results through the parameters, obtained that the Cartesian product interconnection network topology Torus-Butterfly has improved properties of the network topology of existing Cartesian product.
The Senate Committee of Gunadarma University determine these two candidates were graduated with Cum Laude honors.