Real time anomaly classification in surveillance videos on cloud using quantum and deep learning

Abstract

The significance of anomaly detection is essential in the field of intelligent surveillance videos. It is crucial for assessing real-time surveillance footage, including situations like traffic accidents, criminal activities, and violence alerts. CNN, in particular, have historically been the standard of excellence for these applications when it comes to deep learning techniques. Nevertheless, these approaches require significant computational power and resources, resulting in high computational demands. Recognizing the computational limitations, there is a growing enthusiasm for unlocking the capabilities of quantum computing. Quantum computing, an emerging domain, leverages quantum mechanical principles to tackle intricate computational challenges. Consequently, our research aims to transcend the boundaries of conventional computing by introducing an innovative hybrid paradigm, denoted as the Quantum Computing-based Convolutional Neural Network (QC-CNN). This pioneering approach seeks to synergize the power of quantum computing with the neural network framework, paving the way for enhanced problem-solving capabilities at the intersection of deep learning and quantum computation. newline