Blockchain-Based Evidence Authentication Systems: Enhancing Security and Integrity in Criminal Justice
Abstract
To find the definition of smart justice, criminal justice should be effective, fair, and technologically advanced. The national security, fiscal security and population security are highly dependent on the early detection of adverse behavior. However, certain undesirable parts of society are encouraged to commit criminals due to the inequality in the standards of livelihood that may put a strain on the balance of civilization and psychological tranquility. These problems can now be addressed and produce a holistic smart criminal justice system that unpredictably identifies such criminal acts because of the progress in artificial intelligence (AI). The present paper is based on a model of using AI technologies such as deep learning (DL), blockchain, and security strategies to analyze, infer, and propose how criminal justice management can be arranged. Specifically, the architecture offers a secure and inseparable platform to deal with legal documents and operations and preserve data privacy owing to the integration of blockchain. In this study, the authors create a convolution neural network (CNN)-based Xception algorithm named BlockCrime, which is used to detect criminal acts and improve community security. The blockchain innovation securely keeps the known details of criminal activity and alerts the nearest judicial departments. Due to the lack of databases, the transfer learning was chosen, and CNN-oriented Xception approach was used. The Xception new model is evaluated based on a number of evaluation metrics and surpasses the earlier CNN systems in terms of accuracy$(\text{9 6. 6 0 \%} )$. The advantages of the suggested paradigm are in its effectiveness and rapidity, lower risk of getting an error, consistency of applying criminal justice, and additional security and confidentiality.