Lip print, an emerging biometric marker, can be used to identify suspects or victims in criminal investigations and forensic identification. However, problems such as difficulty in feature extraction, slow recognition speed, and low accuracy of lip print images remain. In this paper, we use the latest YOLOv10 image recognition model to identify and classify lip print images, and construct a lip print image dataset based on the CFD face dataset, evaluate the speed and accuracy performance of the differe nt models, and explore and study the deep learning models applicable to lip print recognition. The detection effect of the YOLOv10 model on the lip print image dataset is better than traditional models such as SSD, Faster R-CNN, YOLOv5, YOLOv8, etc. The accuracy rate, recall rate, mAP50 and MAP50-95 of YOLOv10-X model reached 99.3%, 99.9%, 99.5% and 97.4% respectively, but YOLOv10-S and YOLOv10-N had the fastest training time and detection speed, and were more suitable for mobile intelligent devices.