Computational Methods for Video Blockchain in Intelligent Surveillance

This PhD thesis explores how blockchain can enhance intelligent surveillance by improving security and data integrity. It addresses issues such as video tampering, privacy breaches, and security threats in smart cities. The study details how recorded videos are broken into frames and stored within a private blockchain to ensure the immutability of records, boosting trust and security through cryptographic methods. A new concept called Video Blockchain has been introduced to organize surveillance data better. The research demonstrates that video integrity is preserved using the blockchain's tamper-proof features and SHA-256 cryptographic hash functions, while the Merkle tree structure tracks any changes in video frames. This thesis highlights how blockchain has evolved from a cryptocurrency tool into a solution for securing video data in smart cities. It emphasizes blockchain's impact on data safety, particularly in intelligent transportation and surveillance. The private blockchain model used here stores video frames within the blockchain, ensuring secure data transmission across widely dispersed surveillance cameras. More importantly, blockchain ensures the authenticity of surveillance recordings, with potential future applications in artificial intelligence and machine learning. The contributions of this PhD thesis include: • Evaluating Cryptographic Function Performance for Blockchain Implementations. • Implementing Video Blockchain Computation Method for Intelligent Surveillance. • Established a decentralized surveillance system with Video Blockchain. • Addressing Privacy Preservation Issues in Blockchain Implementations • Implementing IoT Video Network Privacy-Secure Mechanisms with Video Blockchain. • Detecting AI-Generated Video Misinformation Using Video Blockchain. Overall, this research is truly significant regarding improved security for surveillance systems, data integrity, and reliability in intelligent surveillance video footage.

Author

• Moolika Gedara, Kasun Tharaka Vishwajith

Identifier

hdl:10292/18151

OAI identifier oai:identifier

oai:openrepository.aut.ac.nz:10292/18151