International Journal of Innovative Science and Modern Engineering (IJISME)
ISSN: 2319-6386, Volume-3 Issue-2, January 2015
57
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: B0788013215/2014©BEIESP
Implementation of an Intelligent Head Gesture
Recognition System
Rushikesh T. Bankar, Suresh S. Salankar
Abstract - As per the rapidly advancement in the changing
technology, the numerous applications are required. For
example, for face, hand and gesture recognition. The previous
researchers have been developed various methods foe head
gesture recognition and they presented various limitations.
Therefore the paper proposes an FPGA based gesture recognition
system. The gesture detection and gesture recognition can achieve
30 frames per second using FPGA system. Accordingly that the
system software can subsequently schedule all the tasks during the
processing. The proposed system also introduces the obstacle
detection technique. The system uses ultrasonic sensors for the
obstacle detection. The proposed system is responsible for the
detection of obstacles.
Index Terms - Wheelchair Interface, Ultrasonic Sensors, Face
Detection, Gesture Recognition.
I. INTRODUCTION
The wheelchair is one of the most commonly method for
enhancing the personal mobility of the users or peoples or the
individuals with the certain disabilities. As per the world
health organization, the total estimated count 1 % of the
world’s population or just over the 65 million people or user
or individuals need a wheelchair. Over the 6.1 million people
in India have movement related disability. The evolution of
the User Interface (UI) witnessed the development from text
based UI based on keyboard to GUI based on mice. In current
virtual environments applications, keyboards, mice, and
joysticks are still the most popular and dominant devices.
However, they are inconvenient and unnatural. The use of
human movements, especially head gestures, has become an
important part of HCII (Human Computer Intelligent
Interaction) in recent years, which serves as a motivating
force for research in modelling, analysing and recognition of
head gestures. Many techniques developed in HCII can be
extended to other areas such as surveillance, robot control and
teleconferencing. Recognizing Gestures is a complex task
which involves many aspects such as motion modelling,
motion analysis, pattern recognition and machine learning.
The successful deployment of the intelligent system relies on
their high performance as well as low cost.
Manuscript Received on January 2015.
Prof. Rushikesh T. Bankar, Asst. Prof., Department of Electronics &
Telecommunication Engineering, G. H. Raisoni College of Engineering,
Nagpur, India.
Dr. Suresh S. Salankar, Prof., Department of Electronics &
Telecommunication Engineering, G. H. Raisoni College of Engineering,
Nagpur, India.
As compared to the other head gesture recognition system, the
main performance of the intelligent system includes the
autonomous navigation capability for good safety, flexibility,
mobility, obstacle avoidance etc., an intelligent system
including voice based control system, vision based control
system, and sensor based control system. The new generation,
an intelligent system of head gesture recognition should be
able to deal with the uncertainties from the practical
applications point of view. They are the user head either out of
the image view or only the profile face is in the captured
image, the face color of the user is user dependent, or may
change dramatically in the varying illumination conditions,
the different facial appearances conditions of the user, such as
mustache and glasses, and the cluttered background. The
proposed system is an intelligent system for the head gesture
recognition, which is based on the combination of Adaboost
and improved camshift algorithms.
II. HUMAN GESTURE REPRESENTATION
There have been many studies on human gestures in
psycholinguistic research. K. Yuan [1] represents the Head
Gesture Recognition for Hands Free Control of An Intelligent
Wheelchair. In this paper, a novel integrated approach to Real
- Time Face Detection, Tracking and Gesture Recognition is
proposed, namely Head Gesture Based Interface (HGI). It is
to be used as the Human Robot Interface for the Intelligent
Wheelchair, namely Robo Chair. The system is used to solve
the problems. They are - The user head may be out of the
image view, or only the profile face is in the captured image,
the face colour is user dependent, and may change
dramatically in varying illumination conditions, the user may
have different facial appearances, such as mustache and
glasses, and the background may be cluttered when an
Intelligent Wheelchair move in the real world. Hyunduk Kim,
Sang Heon Lee, Myoung Kyu Sohn and Dong Ju Kim [2]
present a novel approach for head pose estimation in gray
level images. The two techniques were employed in this
research project. The method of Random Forests was
employed for dealing with the large set of training data. In the
field of computer vision, this is known as a state of art
classification. For the changing in the illumination conditions
that is outdoor environmental conditions, a Binary Pattern
Run Length matrix is very useful. This matrix is combination
of Binary Pattern and a Run Length matrix. The binary pattern
was calculated by randomly selected operator. M. Davy and
R. Deepa [3] present Head Movement System using
Accelerometer Sensor. For detecting the head movements of
the users, the system includes the accelerometer sensors.