Automatic Signature Stability Analysis And Verification Using Local Features

Installation

Clone the repository here.

   >>> git clone https://github.com/swetanjal/Automatic-Signature-Stability-Analysis-And-Verification-Using-Local-Features

Usage

Run Demo (Run Train model before hand)

   >>> python3 demo.py

Train model

   >>> python3 main_train.py

Test model

   >>> python3 main_test.py

Requisites

• opencv-contrib-python: Contains implementation of SURF

   >>> pip3 install opencv-contrib-python

• threading: This code uses threading to parallelize

Folders

• Documents: Research Paper & Project Proposal
• Pickles: Pickle files containing reference database and other intermediate values
• Plots: Contains all generated plots
• src: Code Folder
• TestSet: Test Set of 4NSigComp2010 Dataset
• TrainSet: Train Set of 4NSigComp2010 Dataset

Code Files

• classify.py: Classify matched keypoints
• createDB.py: Create the reference database
• demo.py: Demo for the presentation (Incomplete)
• main_train.py: Generate graphs and results for train set
• main_test.py: Generate graphs and results for test set
• plots.py: Plotting functions

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change. Nishant Sharma, Swetanjal Dutta, Teja Sai Dhondu Please make sure to update tests as appropriate.

Dataset

Dataset and instructions to download can be found here. TrainingSet and TestSet folders must be place in the repository's root folder

License

MIT