Cloud Computing Based Intelligent Video Surveillance Framework for Logistics Security

Logistics services facilitate the worldwide movement of products via interconnected vehicle networks. The flow of commodities is tracked to ensure security, optimization, and safety, among other utilities. Logistics management often uses video surveillance to enhance security by continuously monitoring and identifying events. The imperative requirement for improved security and efficiency in industrial logistics drives this study. A more secure and efficient global logistics industry is essential because of the growing complexity of product transport and tracking via connected vehicle networks. With traditional video surveillance technologies, detecting and classifying events quickly may be challenging. This article proposes an Integral Monitoring and Event Detection Framework (IMEDF) for classifying surveillance security with cloud computing. The framework is used to classify occurrences as either normal or abnormal. The classification depends on various data inputs learned using a convolutional neural network regarding detection and alerting. Logistic monitoring is suggested for optimization and breach mitigation, although it detects anomaly occurrences as a disadvantage. After the categorization process detects normal and atypical occurrences, recommendations are made through intelligent computing in cloud infrastructures. Secure and always-on connectivity are hallmarks of the cloud architecture, which is essential for logistical management. Rate of even detection, time to detection, time to classification, error ratio, and utilization factor are the measures used to verify the performance. An equivalent 15% reduction in classification time accompanied a 25% increase in the event detection rate relative to existing methods. The error ratio dropped 20%, indicating better event classification and suggestion. The utilization factor improved by 30%, indicating better re-source optimization. By employing autonomous training and fine-tuning error-inducing variables, the framework demonstrates a notable enhancement in effectiveness, exhibiting a 25% higher efficiency rate than the current models. Accuracy (93.41), recall (95.67), precision (94.88), and F1 score (96.12) are the metrics that the suggested model possesses. The IMEDF model outperforms the competition in several parameters, including accuracy, precision, recall, and F1 score, which points to its strength in anomaly detection and classification. It should be noted that IMEDF outperformed the alternative models in terms of accuracy, precision, recall, and F1 score.