Automated Signature Verification

Verify the authenticity of signatures by viewing the difference score between two signatures through a web application.

Pre-requisites

This project was implemented using the following libraries in Python 2:

• scipy version 0.18.0
• pillow version 3.0.0
• OpenCV
• theano version 0.9
• Lasagne: install using this link "https://github.com/Lasagne/Lasagne/archive/master.zip"
• web.py
• The pre-trained models can be downloaded from https://drive.google.com/file/d/1KffsnZu8-33wXklsodofw-a-KX6tAsVN/view
• Unzip / extract the contents of the model in "models" folder

How to run the Web Application

• Create a folder and name it anything you want.
• Extract the contents of the master file in this folder.
• Navigate yourself inside the folder that you created.
• Run the following file from command line: "python Web_App/Wisnett_Sig_App.py"
• Open your browser and type localhost in url bar.
• You can upload images of entire documents or cheques and crop the signature part using the "Crop Signature" button.
• Click on calculate features for the signature uploaded first.
• Click on calculate difference score for the singature uploaded second.

Credits

The pre-trained model and parts of the code were adopted from the following repository: https://github.com/luizgh/sigver_wiwd