https://github.com/m-karthik-kumar/eye-blink-counter-using-opencv

It detects the eye blink rate of the user and warns the user if the blink rate is below a set threshold.
https://github.com/m-karthik-kumar/eye-blink-counter-using-opencv

artificial-intelligence machine-learning-algorithms mediapipe neural-networks opencv patent pipelines python3 system-design

Last synced: about 1 month ago
JSON representation

It detects the eye blink rate of the user and warns the user if the blink rate is below a set threshold.

• Host: GitHub
• URL: https://github.com/m-karthik-kumar/eye-blink-counter-using-opencv
• Owner: m-karthik-kumar
• Created: 2023-02-28T09:38:15.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-09-13T23:08:11.000Z (4 months ago)
• Last Synced: 2024-09-14T13:35:01.032Z (4 months ago)
• Topics: artificial-intelligence, machine-learning-algorithms, mediapipe, neural-networks, opencv, patent, pipelines, python3, system-design
• Language: Python
• Homepage:
• Size: 220 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Eye-Blink-Counter-using-OpenCV

 

The eye is considered a specialized light-sensitive organ for gathering visual information. This project proposes a new method for detecting eye blinks without any restriction on the background and the use of any sensors on the eye. Convolution neural network (It is a type of artificial neural network that is used in image recognition and processing which deals with pixel data) is used for recognizing a face and detecting the eyes. To reduce computer vision syndrome (CVS) also known as digital eye strain from which many people suffer nowadays, an eye blink detection system warns the user to stop watching the screen if eyes are getting strained by popping up an alert message. If kids are using the device, the message will be sent to parents if their eyes are getting strained too much, it is detected by the change in eye blinking pattern. The interface is based on a notebook equipped with a typical high-definition web camera and requires no extra light sources in a normal room with sufficient light. This project also proposes a new technique for detecting eye blinks accurately. It automatically classifies the eye as either open or closed at each frame in a video. It also works well on people wearing glasses. Users can even set the timing during which alert notification is to be shown. The eye blink detection system counts the number of blinks every minute and keeps track of the health of the eyes. We use employed image processing methods present in OpenCV to detect the eyeblink action which includes Haar-like features for automatic face detection, and template matching-based eye-tracking and eye-blink detection.