HOMOMORPHIC ENCRYPTION BASED VIDEO COPY DETECTION IN MULTI VIEW VIDEOS STORED IN A CLOUD

Abstract

The rapid growth of cloud computing and multimedia sharing platforms has significantly increased the storage and distribution of large volumes of video data. This has raised serious concerns regarding video piracy, unauthorized duplication, and copyright violations, especially in multi-view video environments where multiple camera angles are involved. Traditional video copy detection techniques often rely on direct feature extraction and comparison, which may expose sensitive content and compromise data privacy. To address these challenges, this project proposes a Homomorphic Encryption Based Video Copy Detection System for Multi-View Videos Stored in Cloud, which ensures both data security and efficient detection. The proposed system utilizes homomorphic encryption, a cryptographic technique that allows computations to be performed directly on encrypted data without requiring decryption. In this framework, video features such as frames, motion vectors, and visual descriptors are extracted and encrypted before being uploaded to the cloud. The cloud server performs similarity detection and comparison operations on encrypted data, ensuring that the original video content remains confidential. This approach enables secure outsourcing of video analysis tasks while preserving user privacy. The methodology involves preprocessing multi-view video data, extracting robust features such as keyframes, color histograms, and spatio-temporal descriptors, and applying homomorphic encryption to these features. A similarity matching algorithm is then used to detect duplicate or near-duplicate videos across different views. The system is evaluated using metrics such as detection accuracy, precision, recall, and computational efficiency. Experimental results demonstrate that the proposed approach effectively detects video copies while maintaining strong data privacy and security. Overall, this system provides a reliable and privacypreserving solution for video copy detection in cloud environments. It is particularly useful for applications in digital rights management, surveillance systems, and multimedia content protection. The integration of homomorphic encryption with video analytics represents a significant advancement in secure cloud-based multimedia processing.