Tagging images of bank cards, such as credit card, debit card, etc, based on Xception pretrained deep feature extraction and my own trained classification layers
Same image tagging examples:
Input | Output |
---|---|
![]() |
{ 'tag': 'bank_card', 'score': 0.9966258 } |
![]() |
{ 'tag': 'non_bank_card', 'score': 0.99998796 } |
to install the bank card detector, you need Python 3.7.7
git clone https://github.com/liang6261515/bank_card_detection.git
cd bank_card_detection
pip3 install -r requirements.txt
wget https://github.com/fchollet/deep-learning-models/releases/download/v0.4/xception_weights_tf_dim_ordering_tf_kernels_notop.h5
download the pretrain model of bank card detection from the following url
https://drive.google.com/file/d/10H70SUi1C5R79cu-27u-bg9ytnZ8JV9F/view?usp=sharing
to use the model to tag the image, we have three cases
Test case #1
the first image looks like
run the test code
from bank_card_detection import bank_card_detection
bank_card_detection('bank_card.jpg')
you will see the following output since the image itself is a credit card
{'tag': 'bank_card', 'score': 0.9966258}
Test case #2
the seconde image looks like
run the test code
bank_card_detection('bank_card1.jpg')
you will see the following output since the image has a group of visa cards
{'tag': 'bank_card', 'score': 0.9973912}
Test case #3
Then we will see one negative casese where the image have no bank card at all. The image looks like
Run the tagging codes
bank_card_detection('non_bank_card.jpeg')
You will see the output of non bank card because there is no bank card at all
{'tag': 'non_bank_card', 'score': 0.99998796}
feel free to contact me if you have any problem with this package or you are hiring data scientist/AI engineer. I am actively looking for data science/AI related jobs
My email: [email protected]