Pilldex: A pill recognition REST API using Convolutional Neural Networks.

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.

• Clone this repo
• Install requirements
• Train the model
• Run the script
• Check http://localhost:5000

• OpenCV
• Keras
• Sqlite3
• CUDA
• Flask

$ git clone https://github.com/ruifgmonteiro/pilldex.git

$ cd pilldex

It is recommended to create a virtual environment to run the application so you can package your application without messing with other python installations.

You just need to give this environment a name and run the following commands:

$ python3 -m venv ./<env_name>

Activate your python environment:

$ source <env_name>/bin/activate

If you want to deactivate it just run

deactivate

pip install .

$ pip install -r requirements.txt

Make sure you have the following installed:

• tensorflow
• keras
• flask
• pillow
• h5py
• gevent
• sqlite3

Python 2.7 or 3.5+ are supported and tested.

$ cd /pilldex/train

$ python train.py --dataset dataset --model pilldex.model --labelbin lb.pickle

$ tensorboard --logdir=Log/ --port=8101

Before running the application you need to setup the database and create the table to store the relevant information.

Change to the actual database directory:

$ cd /pilldex/app/database

Run the script to create both the pills database and table:

$ python create_db.py 

Run the script to insert the data in the table:

$ python insert_db.py

$ python app.py

After the app is running, upload the images you want to classify and check the results!

• OpenCV - Computer Vision Library
• Flask - The web framework used
• Keras - Python high level deep learning library
• Tensorflow - An open source machine learning library for research and production.
• Sqlite3 - Database engine

• Rui Monteiro - Initial work - rfgm6

See also the list of contributors who participated in this project.

This project is licensed under the MIT License - see the LICENSE.md file for details

• Adrian Rosebrock from PyImageSearch

• Building a simple Keras + deep learning REST API
• Deploy Keras Model with Flask as Web App in 10 Minutes