Control_Lights_with_Eyes

CC341 Digital Electronics course project.

Overview

Lights get turned on or off by detecting eye blink using opencv-python computer vision library and dlib machine learning library.

Preview.mp4

Used software

• python 3.7
  
• opencv-python
  
• numpy
  
• dlib
  
• imutils
  
• scipy
  
• serial
  
• Arduino ide

Electronic components

• Arduino board
  
• 2 LEDs
  
• webcam

Circuit.mp4

Commands used

• For detections using webcam video:
  python detect_blinks.py --shape-predictor shape_predictor_68_face_landmarks.dat

• For detections using saved video:
  python detect_blinks.py --shape-predictor shape_predictor_68_face_landmarks.dat --video video_name.mp4

Presentation

controlling lights.pptx

Resources

• PyImage Search:
  https://www.pyimagesearch.com/2017/04/24/eye-blink-detection-opencv-python-dlib/
  
• Real-Time Eye Blink Detection using Facial Landmarks paper:
  https://vision.fe.uni-lj.si/cvww2016/proceedings/papers/05.pdf
  
• Trained landmarks file:
  https://drive.google.com/file/d/1dyMY1F38ByZ9yzpwSSwS8piECkHsBpgQ/view?usp=sharing