Road traffic is a complex phenomenon due to the large number of actors involved and the high density of the network on which it takes place. Consequently, increased traffic volume can lead to various incidents, such as traffic jams or road accidents, which constitute a major problem for society. To combat these incidents and improve road safety, artificial intelligence has proven to be relevant with the implementation of deep learning techniques to improve road traffic monitoring. However, computer vision-based vehicle detection is very effective in providing essential information, such as assessing road traffic conditions and detecting criminal situations. In this paper, we focus on the detection of vehicles traveling on a road in traffic surveillance to provide real-time information to the police or gendarmerie. We therefore propose deep learning based on the identification and detection of vehicles in various traffic surveillance environments. The yolov11 framework was used in this study for real-time vehicle detection and identification. The model used achieved high detection accuracy at low computational cost. YOLOv11 trained on our vehicle detection dataset achieved an average accuracy (mAP) of 90.8%, surpassing existing approaches in terms of detection accuracy.