Chained digital signature for the improved video integrity verification

Abstract

The recorded videos from the surveillance cameras can be used as potential evidence in forensic applications. These videos can be easily manipulated or tampered with video editing tools without leaving visible clues. Hence integrity verification is essential before using the videos as evidence. Existing methods mostly depend on the analysis of video data stream and video container for tampering detection. This paper discusses an active video integrity verification method using Elliptic Curve Cryptography and blockchain. The method uses Elliptic Curve Digital Signature Algorithm for calculating digital signature for video content and previous block. The digital signature of the encoded video segment (video content with predetermined size) and that of previous block are kept in each block to form an unbreakable chain. Our method does not consider any coding or compression artifacts of the video file and can be used on any video type and is tested on public-available standard videos with varying sizes and types. The proposed integrity verification scheme has better detection capabilities towards different types of alterations like insertion, copy-paste and deletion and can detect any type of forgery. This method is faster and more resistant to brute force and collision attacks in comparison to existing recent blockchain method.