Smart Border Surveillance System Based on Deep Learning Methods

Any country has to protect its borders in order to maintain peace and ensure the safety of the people. To control illegal movement of beings and terrorist attacks, it becomes vital to use advanced smart technology such as deep learning. It has recently gained power in smart cities and societies, such as in healthcare, surveillance, and in a variety of artificial intelligence-based real-life applications. Border surveillance is a big challenging task in modern computer vision applications. One of its potentials is object detection. It has achieved auspicious results concerning the finding of an object in images. The objective of this paper is to present a smart border surveillance system to monitor doubtful activities across the border or nearby red zones and help in controlling dangerous situations. It annotates and localizes persons, animals and planes in images detected by cameras in order to supervise the borders of a country. A comparison between three algorithms is presented in this paper. Experimental results reveal that YOLOv5 outperforms the others in terms of speed and accuracy.