<![CDATA[The rapid development of urbanization and security concerns has prompted the integration of automated smart surveillance systems to enhance public safety and operational efficiency. Traditional surveillance methods often rely on human monitoring, which is prone to errors and inefficiencies. Image processing and computer vision techniques provide a solution by automating object detection, tracking, and anomaly recognition. This study aims to investigate advanced image processing and computer vision techniques for improving the performance of automated smart surveillance systems. A hybrid approach combining convolutional neural networks (CNNs), attention mechanisms, and edge computing is proposed to enhance both detection accuracy and real-time processing speed. The research employed experimental design, utilizing a dataset of 12,000 annotated image frames and 85 hours of video footage from diverse environmental conditions. Performance metrics such as precision, recall, mean average precision (mAP), and processing speed were measured. Results demonstrate that the proposed model outperforms traditional CNN models, achieving higher detection accuracy and faster processing speed. The study concludes that integrating edge computing with adaptive image processing and attention-based neural networks significantly improves automated surveillance system performance in real-world settings. These findings offer valuable insights for the development of scalable and efficient smart surveillance technologies.]]>
Copyrights © 2025
