节点文献
一种基于机器视觉的机场场面多目标跟踪算法
A Multi Target Tracking Algorithm for Airport Scene Based on Machine Vision
【摘要】 阐述YOLOv5s和DeepSORT改进的多目标跟踪算法,YOLOv5s作为前置的目标检测器,使用自制的数据集进行模型训练,探讨随机调整图像的对比度参数提升模型的泛化能力。
【Abstract】 This paper describes the improved multi-target tracking algorithm of YOLOv5s and DeepSORT. YOLOv5s, as the pre target detector, uses the self-made data set for model training, and discusses how to randomly adjust the contrast parameters of the image to improve the generalization ability of the model.
【关键词】 YOLOv5s;
DeepSORT;
模型量化;
多目标跟踪;
【Key words】 YOLOv5s; DeepSORT; model quantification; multi-target tracking;
【Key words】 YOLOv5s; DeepSORT; model quantification; multi-target tracking;
- 【文献出处】 电子技术 ,Electronic Technology , 编辑部邮箱 ,2022年12期
- 【分类号】TP391.41;V35
- 【下载频次】20