New PDF Forgery Detection Technique Analyzes File’s Page Objects

Researchers from the University of Pretoria have developed a new method to detect tampering or forgery in PDF documents. The technique focuses on analyzing the page objects of a PDF, which includes text, images, and metadata. The challenge with current techniques like watermarking and hashing is that they only identify alterations to visible content, without detecting changes to hidden elements such as metadata or embedded malware. The new prototype, created using Python and libraries like hashlib, Merkly, and PDFRW, creates unique digital fingerprints for each page’s content and stores these hashes in a hidden, unalterable record. This allows the document to be ‘protected’ and later checked for changes. When a document is checked for forgery, the prototype compares the current document’s content with the stored hash values, pinpointing the exact alterations down to specific sections of the document. While effective for detecting changes made with Adobe Acrobat, the prototype cannot yet detect all types of changes, such as font alterations or JavaScript code modifications. Despite its limitations, the prototype offers a promising solution for protecting PDF documents from tampering and ensuring the integrity of digital communication.