Fraudulent Face Image Detection

Ramrao Adik Institute of Technology, Dept. of Information Technology, Mumbai, India

^* e-mail: rahul.awhad28@gmail.com
^** e-mail: srsjayswal@gmail.com
^*** e-mail: adeshmore98@gmail.com
^**** e-mail: jyoti.jadhav@rait.ac.in

Abstract

Due to the growing advancements in technology, many software applications are being developed to modify and edit images. Such software can be used to alter images. Nowadays, an altered image is so realistic that it becomes too difficult for a person to identify whether the image is fake or real. Such software applications can be used to alter the image of a person’s face also. So, it becomes very difficult to identify whether the image of the face is real or not. Our proposed system is used to identify whether the image of a face is fake or real. The proposed system makes use of machine learning. The system makes use of a convolution neural network and support vector classifier. Both these machine learning models are trained using real as well as fake images. Both these trained models will take an image as an input and will determine whether the image is fake or real.