Вы находитесь на странице: 1из 2

Handsfree Mouse Control Using Eye Movement

Prof. Premanand Ghadekar, Pragya Korpal, Pooja Chendake ,Raksha Bansal, Apurva
Pawar,Siddhi Bhor
Department of Information Technology ,
Vishwakarma Institute Of Technology ,Pune,India.

---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract – -In this paper, an individual human
computer interface system using eye motion is Design and Implementation
introduced. Traditionally, human computer interface The proposed system which will accomplish our
uses mouse, keyboard as an input device. This paper objectives for projects are defined in those below
presents hands free interface between computer and stages of figure system flow.
human. The system we describe is real time, on- A. Start: Whenever the user will Start the next module
intrusive, fast and affordable technique for tracking will open.
facial features. This technology is intended to replace Stop: Stop will get the user out of the application.
the conventional computer screen pointing devices for When user will run the module the camera will open
the use of disabled. The paper presents a novel idea to automatically.
control computer mouse cursor movement with human B. Recording Videos and Convert it in Frames:
eyes It controls mouse-moving by automatically Whenever the camera will open it will start recording
affecting the position where eyesight focuses on, and the video. The videos will record the movement of user.
simulates mouse-click by affecting blinking action. The recorded videos are now converted into number of
However, the proposed vision-based virtual interface frames. The frames recorded per second would be
controls system work on various eye movements such 100fpm .
as eye blinking. C. Converting RGB into grey scale: When the videos are
converted into frames they are in RGB format. These
Keywords – Eye tracking, mouse movement, Eye- frames which are in RGB format now converted into the
blinking detection. Grey scale format.
D. Eye Detection & Tracking: As, there are only two
Proposed System colors it will easy to detect the users eye. After the
Our system is real time which captures a movement of conversion of frames into grey-scale the eye will be in
mouse cursor through face detection and facial white color.
features. It overcomes the existing system by avoiding Now, the eye detection approaches will be applied on
the use of external hardware that caused serious eye the frames to detect eye. Then the system will detect
damages. It uses a template matching method for eye the eye it will track the movement of the eye.
extraction instead of using hardware, even as in After these all above steps are done successfully basic
previous system the short blinks of eyes were avoided function such as moving the cursor would be
or neglected. In this system the hard blink is only used performed.
for selecting particular file or folder .With eye detection
it’s first aim is to captured face for the movement of
mouse cursor. Then it reacts as the mouse does.
Start

Recording Videos and


convert it in frames

Converting RGB into


grey scale

Eye Detection and


Tracking

Stop Basic Operation

Dataset Used
We have used shape_predictor_68_face_landmarks
dataset. The pose takes the form of 68 landmarks.
These are points on the face such as the corners of
the mouth, along the eyebrows, on the eyes, and so
forth. The face detector we use is made using the
classic Histogram of Oriented Gradients (HOG)
feature combined with a linear classifier, an image
pyramid, and sliding window detection scheme.

Вам также может понравиться