Computer Vision-Based Military Tank Recognition Using Object Detection Technique: An application of the YOLO Framework

• Published in: 2023 1st International Conference on Advanced Innovations in Smart Cities (ICAISC) pp. 1 - 6
• Main Authors: Ali, Sikandar, Abdullah, Athar, Ali, Ali, Maisam, Hussain, Ali, Kim, Hee-Cheol
• Format: Conference Proceeding
• Language: English
• Published: IEEE 23-01-2023
• Subjects: Computational modeling; Computer vision; Deep learning; Military computing; Military Tanks detection; Object detection; Smart cities; Surveillance; Technological innovation; Training; YOLOv5
• DOI: 10.1109/ICAISC56366.2023.10085552

Military object detection is an indispensable and challenging task for defence systems which includes the tracking, tracing, security, and surveillance of any territory or region. These systems should be very efficient, reliable, and accurate in executing their functions. A minute errant may result in mass destruction and loss. So automatic real-time object detections are imperative in today's world. Although over the years, different traditional approaches and techniques have been used for the detection of military equipment, warheads, and other defence-related objects yet the efficiency and accuracy of those techniques are comparatively low compared to the artificial intelligence-based object detection techniques. Therefore, we demonstrate the latest computer vision-based real-time object detection technique to detect real-time military objects with high accuracy and precision. We introduced YOLOv5 for the detection of military tanks and flags. This model successfully detects the targeted objects i.e., tank and flag with high confidence and precision. We trained and evaluated the performance of YOLOv3, YOLOv4, and four versions of the YOLOv5 model i.e., YOLOVv5s, YOLOv5m, YOLOV51, YOLOV5x1 with 922 images consisting of tank and flag objects. The dataset has been divided into 80% training, 10% validation, and 10% testing. The detection results of all six YOLO versions are compared and evaluated. The experimental results showed that the YOLOv5xl achieved higher performance. The precision, recall, mAP_0.5, and mAP_0.95 were 0.99, 0.995, 0.995, and 0.892, respectively. Since YOLOv5 is one of the latest and fastest real-time object detection approaches so this model will empower and enhance the military surveillance systems by enabling the military personnels to take prompt and proactive actions against any potential threats.