covid19-detection

Detection of Covid-19 from X-ray images

Status:

Train

python3 TRAIN.py -d dataset/test -m model_name.hdf5

NOTE :

• Model with model_name.hdf5 will save in SavedModel directory.
• ep variable in TRAIN.py file is epoch count. you can change it.

example:

python3 TRAIN.py -d dataset/test -m amin.hdf5

result:

Test

test 10 random image

python3 test_model_10_images.py -d dataset/validation -m model_name.hdf5

load and show 10 labeled samples

example:

test single image

python3 test_model_1_image.py -i path_to_image  -m path_to_model

example:

python3 test_model_1_image.py -i ./dataset/one/covid/Chest.jpeg  -m ./SavedModel/amin.hdf5

API ( WIP )

for gathering more images and make the dataset better, I create e simple API for upload the X-RAY image like below examples: The uploaded image will validate after upload and the server return label of the image. Label my be incorrect. (because of the low count of images in the dataset).

API Docs

run on your host

• add mongodb address to conf.env file
• run sudo docker-compose up -d
• open docs http://host:8838/docs

Website Version

• Work In Progress

Flutter Version

• Flutter Version will develope by @EhsanTgv