Finger-Detection

Project Structure

The software used for building the finger detection system is organized as follows:

• data
  • Training-Dataset : Train Images-Masks and Augmented Images
  • Validation-Dataset : Valid Images-Masks and Augmented Images
• preprocessing
  • skin_color_histogram.ipynb : Color spaces analysis
  • skin_color_characterization.ipynb : Skin detection
• models
  • components.ipynb : Finger classification using morphological operators
  • data_augmentation.ipynb : Image augmentation
  • cnn.ipynb : Finger classification using Convolutional NN
• demo
  • finger_detection.py : Real Time System

Pipeline Exectuion

To allow third parties to reproduce the results, execute all the script in the following order:

1. skin_color_histogram.ipynb
2. skin_color_characterization.ipynb
3. components.ipynb

Alternatively, one can execute the finger_detection.py script and the detection & clasification system will pop-up in your laptop:

pyhton3 finger_detection.py [CAMERA SOURCE]

The CAMERA SOURCE is 0 for laptop-cam or the corresponding port for a web-cam, default 1. The default configuration of the system perform at its best if the hand is shown at a distance of 0.4-0.6 meters. w.r.t the camera.

Notice that if cnn.ipynb is going to be executed, it is strongly recommended to make us of GPU's power as well as execute data_augmentatio.ipynb to generate more images.

Requirements

All the work has been done using several libraries, one can install all of them executing:

pip install -r requirements.txt