Person-Identification-Using-EEG

A python based model for identifying a person from a group of people using his/her raw EEG signals

Here in this project we have tried to identify a person from a group of people using his/her raw EEG signals which we have in the form .vhdr files. We first did classification of EEG signals using different deep learning models like Autoencoders, RNN and CNN in order to find which model gives best performance while dealing with EEG signals and we found that CNN model was the best performer with an accuracy of more than 80%. So we decided to use CNN model for person identification task. The person that we wanted to identify was assigned a 1 label and all other people were assigned 0 label. Due to limited resources in terms of RAM and computing power, we used only 5 persons' data in our project. After training the CNN model on EEG data of 5 persons, we evaluated our model and found that our model was able to identify the target person with an accuracy of about 99%. So we can say that the CNN model can be used as a means of identifying people using their EEG signals with a very high accuracy.

The research paper that we used was: https://www.sciencedirect.com/science/article/abs/pii/S0925231220309760

I have also attached 4 data files for your reference and you can use them for training and evaluating the model.

However if you wish to work with more data files, then here's the link to dataset: https://ftp.gwdg.de/pub/misc/MPI-Leipzig_Mind-Brain-Body-LEMON/EEG_MPILMBB_LEMON/EEG_Raw_BIDS_ID/