Project: EyeHelp

https://www.youtube.com/watch?v=48XsWaWJveQ&ab_channel=BCI-NERD

Problem:-

Eye movement disorder is a medical condition characterized by abnormalities in the movement of the eyes, which can affect an individual's ability to track moving objects, read, and perform daily activities that require visual attention. This disorder can manifest in various ways, including involuntary eye movements, difficulty focusing on objects, or problems with eye coordination.

Solution:-

Virtual Keyboard with Eye Tracking

This is a virtual keyboard application that utilizes eye tracking to enable typing without physical input devices. It allows users to control the keyboard and type by blinking their eyes.

SetUp and Installation

To run this application, you need to have Python and the following libraries installed:

1. Clone the repository:

   git clone <repository-url>
   

2. Install the following dependencies.

• OpenCV (cv2)
• math (hypot)
• NumPy (numpy)
• Dlib (dlib)
• Pyglet (pyglet)

3. You can install these dependencies using the following command:

   pip install requirements.txt
   

4. Download the pre-trained facial landmarks model file from here: Link

5. Run the script by executing the following command:

   python main.py 
   

The application will start and open the webcam feed. Place your face in front of the camera and make sure it is properly detected. You will see a virtual keyboard displayed on the screen.

To control the keyboard, follow these instructions:

• You can select the left or right side of the keyboard by gazing at the corresponding side for a few seconds. You will hear a sound left or right depending on the side you select. Once that is selected, it will display the gaze side virtual keyboard.
• To type a letter, blink your eyes. The active letter on the keyboard will be added to the text input area or frame.
• The text you type will be displayed in the text input area at the top of the screen.

I have done two main tasks to solve the conditions of eye movement disorder

1. Eyeblink detection

Eyeblink detection refers to the process of identifying and quantifying the occurrence of blinks in a person's eye movements.

Blinking occurs when:-

• Eyelids are closed
• Eyeballs are no longer visible
• The bottom and upper eyelashes connect

Keynotes:

• In both closed-eye and open-eye, the horizontal line is almost the same size.
• Compared to the closed eye, the vertical line in the open eye is much longer.
• The vertical line almost disappears when the eye is closed

2. EyeGaze Detection

Eyegaze detection, also known as eye tracking, is a technology that enables the tracking and analysis of eye movements and gaze direction. Eyegaze detection systems typically use cameras and infrared sensors to capture eye movements, and sophisticated algorithms to analyze the data. The main goal of eyegaze detection is to determine where a person is looking, which can provide valuable insights into their visual attention and cognitive processes. It has various applications in different fields, including human-computer interaction, psychology, neuroscience, market research, and assistive technology.

• We can easily track both eyes by using gaze detection mechanism as shown in the above figure.