Abstract: Advancements in technology and globalization have made digital cameras widely accessible and affordable. As a result, people capture and collect numerous images using various camera sensors, often in soft copy for online documents and sharing on social media. Following the explosive use of social networking services, there has been an exponential increase in the volume of data of image. Moreover, the development of image processing software such as Adobe Photoshop has given rise to doctored images. These manipulated images can be used for malicious purposes, such as spreading false information or inciting violence. This image spam detection program allows users to detect even the smallest signs of fraud in images. With the rise in crime, image fraud has become a major problem that needs attention.
Moreover, the main goal of forgery detection in the digital age is to ensure immaculacy and validity. As research progresses, many deep learning methods are being implemented to identify fraud in images. Deep learning approaches have shown much better results for image manipulation compared to traditional methods. In this study, we have also aimed to determine the detection of image forgery using a deep learning approach. We propose a novel image forgery detection system based on Convolutional Neural Networks (CNNs) that can detect various types of image modifications, such as copy-move, splicing, and resampling. Our proposed system integrates Error Level Analysis (ELA) with deep learning techniques to provide an accuracy of 93% for detected images. Our proposed system even integrates Visual Geometry Group; it is a standard deep Convolutional Neural Network (CNN) architecture with multiple layers. After evaluating the proposed system on a database of real-world images and achieving a high detection VGG16's training accuracy of 93.21% and a training accuracy of 95.12% for VGG19. VGG16 is the first VGG network and VGG19 is the last hence we decided to use both of them as they give better accuracy than any other networks.
Keywords: digital cameras, doctored images, malicious purposes, authenticity, integrity, image forgery, deep learning, CNN, ELA, VGG16, VGG19
Cite:
Prof. Swati Dronkar, Mauli Bhalotiya, Yashshree Bidkar, Mayank Futane, Umesh Amru, "Image forgery Detection", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 1, 2024, Crossref https://doi.org/10.17148/IJARCCE.2024.13113.
|
DOI:
10.17148/IJARCCE.2024.13113
