This study presents a truck classification method using panoramic side-view images to meet the Ministry of Land, Infrastructure and Transport’s 12-category standard (types 4–12). The system captures a vehicle’s full side profile via a panoramic imaging device, ensuring complete wheel visibility. A YOLOv12-based deep learning model detects wheels, and image processing extracts their center coordinates. Pixel distances between adjacent wheels are calculated and normalized to determine axle spacing patterns, which, together with wheel count, are applied to a rule-based classifier. Tests on 1,200 real-world panoramic truck images (1,000 for training, 200 for testing) achieved a mean average precision of 96.1% for wheel detection and 90.5% overall classification accuracy. The method offers explainable classification through measurable structural features, supporting applications in smart tolling, road usage billing, overloading enforcement, and autonomous vehicle perception.

Overloaded and improperly loaded trucks cause serious road hazards, such as rollovers and cargo falls. Although automatic enforcement methods are being studied, they face challenges in accuracy and legal application. Thus, a technology for direct tracking and enforcement is needed. This study uses EfficientNet to extract features of vehicles and license plates, and applies cosine similarity to identify the same vehicle. Comparisons were divided into “same vehicle” and “similar vehicle,” with a threshold-based method and five classification types. Results showed that the average similarity of the same vehicle group was 0.11 higher than that of the similar vehicle group. The accuracy of correctly identifying the same vehicle was 84.54%. Integrating OCR or LPR is expected to further improve tracking performance.

The lane designation and the bus-only lane system for traffic speed and road safety are difficult to crack down on, and for this purpose, crackdown methods using image recognition technologies are being studied. Existing studies require continuous learning or additional equipment, and it is difficult to classify combined vehicles such as vans and pickup trucks. Therefore, in this study, YOLO and EasyOCR were mixed to classify combined vehicles through vehicle type symbols. For combined vehicles, higher accuracy was shown than classification using YOLO. Due to the nature of Hangul, the accuracy was slightly lowered because the OCR was not accurately recognized, but if it is used with the existing YOLO classification, high accuracy of crackdown will be possible.

The lane departure warning device can not detect the lane to be driven in the future by sensing the departure of the lane passing by during driving and warning the driver. Considering the safe operation of the truck, it is also expected that the departure of the future lanes according to the dynamic weight and speed of the current truck should be predicted. This study attempted to predict whether or not to deviate from the lanes of curved roads to be driven in the future according to the current dynamic driving weight and speed in consideration of the safe driving of trucks.

Recently, traffic accidents have continued to occur due to the failure to secure a safe distance for trucks. Unlike passenger cars, freight cars have a large fluctuation in the weight of the vehicle's shaft depending on the load, and the fatality of accidents and the possibility of accidents are high. In this study, a braking distance prediction model according to the driving speed and loading weight of a three-axis truck was implemented to prevent a forward collision accident. Learning data was generated based on simulation, and a prediction model based on machine learning was implemented to finally verify accuracy. The extra trees algorithm was selected based on the most frequently used R2 Score among regression analyses, and the accuracy of the braking distance prediction model was 98.065% through 10 random scenarios.