Automatic Identification of Truck Head and Tail Using YOLOv10
Systems Science & Control Engineering, (2026), Vol. 14, No. 1
Meijue Xu
a,b
,
Alexander A. Hernandez
c
,
Kaijie Zhang
d
,
Chao Wang
e
a School of Information and Artificial Intelligence, Anhui Business College
b College of Industrial Education, Technological University of the Philippines
c College of Computer Studies and Multimedia Arts, FEU Institute of Technology
d School of Informatics and Engineering, Suzhou Aviation Vocational University
e School of Artificial Intelligence and Big Data, Suzhou University
Abstract: Vehicle head and tail recognition systems play an important role in the monitoring and management of vehicles transporting mineral resources in smart mining systems. This work evaluates the accuracy of head and tail recognition using three versions of the You Only Look Once (YOLO) algorithm series: YOLOv5, YOLOv9, and YOLOv10. In the same experimental environment, by analyzing and comparing the recognition rate, parameter quantity, and recognition accuracy of the object detection model during the experimental process, the Sota model was selected before 2024. The experimental results showed that the recognition accuracies of YOLOv5, YOLOv9, and YOLOv10 were 73.93%, 91.80%, and 99.13%, respectively. Experiments show that under the same equipment conditions, the YOLOv model for identifying and detecting the head and tail of vehicles in the same dataset performs better, which has more advantages in improving the transportation of smart mines.