SpliceFound: A Web-Based Image Splicing Detection Tool Using Error Level Analysis (ELA)

Abstract

Online image sharing is common nowadays, which people share images from social media to digital communication tools. However, the increasing number of modified contents in today's fast-paced digital environment makes it difficult for users to trust the authenticity of images found or received. Moreover, most existing image forgery detection tools are not easy to use, as their outputs can be hard to understand without specialized knowledge and do not provide clear visual indicators or explanations. Furthermore, they usually lack functionalities such as downloadable reports or user accounts, which complicates the process for users trying to monitor or access previous analyses. Furthermore, a lot of them needs complicated configurations, which discourages users without technical expertise from using them efficiently. This led to the creation of SpliceFound, a forensic image analysis tool that detects tampered areas specifically images splicing forgery. SpliceFound, which is a web-based tool, integrates Error Level Analysis (ELA) with thresholding and image masking to provide tampered regions while allowing user account creation and providing additional report on the analysis result. Agile methodology is conducted in the project which emphasizes collaboration, feedback, and adaptability for a user-centered product by the phases of planning, designing, developing, testing, and delivering in iterative cycles. The tool is developed using Python for backend, HTML, CSS with JavaScript for frontend and flask framework. SpliceFound aims to offer a user-friendly image splicing detection tool that provides reasonably accurate results, an automated tampering result reports, and accessible features for forensic practitioners, legal officers and non-technical users. Additionally, it offers the ability to track the history of previous analysis.