Ng, Ing Kye (2014) A real-time vision-based touchless interaction of virtual heart using integration of motion gesture interaction methods. Universiti Malaysia Sabah. (Unpublished)
|
Text
a real time vision based touchless.pdf Download (4MB) | Preview |
Abstract
"Motion gesture is one of the gesture recognition algorithms which provide touchless interaction within a system. The main objective of this project is to design a system that detects motion gesture of a virtual heart in medical imaging environment. This project encompassesa few gesture algorithmss uch as GestureE stimation,F eedback and Gesture Recognition. The purpose of Gesture Estimation is to detect gestures and locates the screen coordinates. The Feedback is to produce response from the user regarding the motion data and the position of gestures. Gesture Recognition is to recognise the gestures and performs the basic transformation of the virtual heart. A Predefine Offset Screen Distance for Gesture Recognition is proposed. The flow of the algorithm can be divided into four steps. The first step is to calculate the offset line. Second, Recognition Threshold is declared to diminish the noise of the motion. Third, the range which is defined as left or right is used for gesture recognition. Lastly, the basic transformation such as rotation, translation and scale is executed if a data motion lies within the range. Four experiments are carried out to evaluate the prototype system. The experiments are Gesture Test, Lighting Test, Distance Test and Memory test. Besides, the proposed Predefine Offset Screen Distance is further analysed by measuring time complexity using Big-O Notation. The results showed that the proposed Predefine Offset Screen Distance is not very efficient and it has higher computational cost in time complexity with 0(n2). In this study, a prototype of Augmented Reality system that implements gesture motion is developed. The use of the gesture motion algorithm enables an interactive environment in medical imaging.
| Item Type: | Academic Exercise |
|---|---|
| Keyword: | Motion gesture, algorithm, Big-O Notation |
| Subjects: | Q Science > Q Science (General) |
| Department: | FACULTY > Faculty of Science and Natural Resources |
| Depositing User: | NORAINI LABUK - |
| Date Deposited: | 21 Nov 2017 13:46 |
| Last Modified: | 21 Nov 2017 13:46 |
| URI: | https://eprints.ums.edu.my/id/eprint/17532 |
Actions (login required)
 |
View Item |