Skip to content

Latest commit

 

History

History
26 lines (19 loc) · 972 Bytes

README.md

File metadata and controls

26 lines (19 loc) · 972 Bytes

Hand Number Recognition

This repo contains script for training and testing Hand Number Recognition. We collect dataset manually using usb webcam and split it into images. Training the model is done using Keras with Tensorflow backend.

Requirements

  • Python 3.6
  • Keras 2.4
  • Tensorflow 1.3
  • Scikit-learn
  • OpenCV 3
  • Imutils
  • Argparse

Download the model

You can download the dataset here

Testing

For testing model using webcam with parameters -m for models, -l for label and -i for input video.

$ python src/classify_video.py -m models/lenet_64x64/hnr_lenet_150x150.model -l models/lenet_64x64/lb.pickle -i videos.mp4

Credits

1. Pyimagesearch Image Classification with Keras and Deep Learning

Contact

If you have questions please email me at [email protected].