• Saikrishnan
• Karthikeyan

Video surveillance performs an essential function in maintaining security and situational awareness through different fields, comprising critical infrastructure, smart cities, and public spaces. The requirement for effectual and robust object detection and classification techniques in video surveillance has considerably developed. This study develops a Selfish Herd Optimizer with deep learning Driven Object detection and Classification Model (SHODL-ODC) for Video Surveillance Systems. The main aim of the SHODL-ODC model is to recognize and classify the existence of objects in the surveillance videos. YOLO-v5 works as the backbone of our model, confirming fast object detection in the surveillance video. For improving the speed and performance of the model, we implement the Nadam optimizer for hyperparameter tuning. Additionally, we present an innovative method for object classification employing Multi-Class Support Vector Machines (MSVM). MSVMs are recognized for their capability to effectively manage multi-class classification tasks, ensuring accurate categorization of objects identified by YOLO-v5. To fine-tune the model's parameters and enhance its overall performance, we developed the Selfish Herd Optimizer (SHO). With comprehensive experimentation and assessment, we establish the model's effectiveness in real-time video surveillance conditions, achieving computational efficiency, classification performance, and higher object detection accuracy. The comparison study represented the improved simulated outcomes of the SHODL-ODC model over the other techniques.

Video surveillance; Object detection; Deep learning; Object classification; Parameter tuning; YOLO-v5

"Selfish Herd Optimizer with Deep Learning Driven Object detection and Classification Model for Video Surveillance Systems", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 9, page no.g134-g145, September-2023, Available :http://www.jetir.org/papers/JETIR2309634.pdf

"Selfish Herd Optimizer with Deep Learning Driven Object detection and Classification Model for Video Surveillance Systems", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 9, page no. ppg134-g145, September-2023, Available at : http://www.jetir.org/papers/JETIR2309634.pdf