This repository contains the code and data sets used for developing a system that verifies an image of an individual's signature as genuine or forged.
Details of the files are as follows:
- trainer.py - Trains weights for the CNN-based feature extractor.
- cnn_ks.py - Performs verification for a set of test signatures using CNN-based feature extractor.
- dct_ks.py - Performs verification for a set of test signatures using DCT-based features.
- network.py - Contains the architecture of the CNN used for feature extraction.
- Signature_Verification.pdf - Deatiled report of the project.