Detection and Localization of Multiple Image Splicing Using MobileNet V1

Abstract

In modern society, digital images have become a prominent source of information and medium of communication. The easy availability of image-altering softwares have greatly reduced the expenses and expertise required to exploit visual tampering. Images can, however, be simply altered using these freely available image editing softwares. Two or more images are combined to generate a new image that can transmit information across social media platforms to influence the people in the society. This information may have both positive and negative consequences. Hence there is a need to develop a technique that will detect and locate a multiple image splicing forgery in an image. This research work proposes multiple image splicing forgery detection using Mask R-CNN, with a backbone as a MobileNet V1. It also calculates the percentage score of a forged region of multiple spliced images. The comparative analysis of the proposed work with the variants of ResNet is performed. The proposed model is trained and tested using the MISD (Multiple Image Splicing dataset), and it is observed that the proposed model outperforms the variants of ResNet models (ResNet 51,101 and 151). The proposed model achieves an average precision of 82% on Multiple Image Splicing Dataset, 74% on CASIA 1.0, 81% on WildWeb, and 86% on Columbia Gray. The F1-Score of the proposed method on MISD was 67%, 64% on CASIA 1.0 68% on WildWeb, and 61% on Columbia Gray, outperforming ResNet variants.