Moving Object Detection Based on Enhanced Yolo-V2 Model
Mukaram Safaldin, Nizar Zaghden, Mahmoud Mejdoub
Abstract
Object detection is a crucial aspect of computer vision, and deep learning has led to significant advancements in this area. Object categorization, segmentation, and localization have been improved with the use of deep models. Two-stage detectors have higher identification precision, while single-stage detectors have better inference times. You Only Look Once (YOLO) and its successors have shown improved detection accuracy and speed, making them popular in a wide range of applications. This paper proposes an improved YOLO-v2 for detecting tiny objects. The proposed detector is evaluated using the VOC 2012 benchmark dataset, and the experimental results show that it outperforms state-of-the-art detectors in terms of detection accuracy, precision, recall, and IOU. The proposed detector achieved 95.8% detection accuracy, 96.1% precision, 95.5% recall, and 95% IOU.