Gesture Recognition Based On HMM With The Signal From INS

Melani, Roro Inda (2011) Gesture Recognition Based On HMM With The Signal From INS. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 2209205017-Master_Thesis.pdf Restricted to Repository staff only Download (8MB) | Request a copy

Abstract

The aim of this project is to recognize human hand gesture through the use of an INS sensor with the approach of a statistical pattern recognition tool, called HMM (Hidden Markov Model). It is assumed that a hand gesture could be detected by its fluctuation in acceleration and the difference of hand-position in 3-axis. Therefore this project employed an INS sensor to receive the input of human hand gesture. This sensor came with 6 sensor channels to generate signals of a 3-axis gyroscope and a 3-axis accelerometer. The acceleration fluctuated in three perpendicular directions due to different hand gestures detected by the accelerometer. And the change of hand-position in 3-axis is detected by gyroscope. The sensor data exported to computer via USB port. During the stage of data collection a cut algorithm was developed to pick the most significant part of the sensor data. After comparing the data, DCT selected to change the signal from time domain to frequency domain. Sequences of calculation were performed to analyze the best sampling frequency to select dominant frequency of every gesture to be picked as parameter value. The parameter value used in HMM as the approach to recognize and differs gestures.

Item Type:	Thesis (Masters)
Additional Information:	RTE 004.019 Mel g 2012 3100012047150 (WEEDING)
Uncontrolled Keywords:	gesture, hand gesture recognition, INS sensor, 3-axis accelerometer, gyroscope, HMM
Subjects:	T Technology > T Technology (General) > T59.7 Human-machine systems. T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing.
Divisions:	Faculty of Industrial Technology > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Anis Wulandari
Date Deposited:	20 Oct 2026 03:03
Last Modified:	20 Oct 2026 03:03
URI:	http://repository.its.ac.id/id/eprint/145583

Actions (login required)

View Item