To optimize the balance between real-time performance and detection precision, the open-sourced AI machine learning tool, YOLOXs: YOLOX-s and YOLOX-nano, were trained and evaluated. The Intersection over Union (IoU) and confidence threshold parameters were tuned within a range of 0.4 to 0.7 to maximize the mean average precision (mAP) of the YOLOX models (see Figure S4).