Recognition of Front-Vehicle Taillights Based on YOLOv5s

Abstract

In automatic driving, the recognition of Front-Vehicle taillights plays a key role in predicting the intentions of the vehicle ahead. In order to accurately identify the Front-Vehicle taillights, we first analyze the different characteristics of the vehicle taillight signal, and then propose an improved taillight recognition model based on YOLOv5s. First, CA(coordinate attention) is inserted into the backbone network of YOLOv5s model to improve small target recognition and reduce interference from other light sources; Then, the EIOU Loss is used to solve the class imbalance problem; Finally, EIOU-NMS is used to solve the problem of anchor box error suppression in the recognition process. We use the actual scene video and vehicle taillights dataset to conduct ablation experiments to verify the effectiveness of the improved algorithm. The experimental results show that the mAP value of the model is 9.2% higher than YOLOv5s.INDEX TERMS Autonomous driving, vehicle taillights recognition, ablation experiment.

Review of Integrated Chassis Control Techniques for Automated Ground Vehicles

“…The perception methods such as object classification, object tracking, and vehicle localisations make sense of raw data that are brought after the pre-processing step by sensors and V2X communication. AD and ICC can use cameras to estimate velocity using visual odometry, yaw, roll, and pitch rates, detect road lines [ 149 , 150 ], road curve angle and road bank angle, intersections, railroad, and pedestrian crossings, road surface type and conditions [ 118 ], road lanes and boundaries [ 151 , 152 ], road signs including warnings and speed limits [ 153 ], and horizontal marking detection and recognition [ 154 ], road damage, potholes, and distress detection [ 155 ], location [ 156 ], velocity and displacement of the vehicle [ 157 ] and other vehicles on the road even using their taillights [ 158 ]. ICC systems can use previously described advanced sensors to acquire horizon prediction, and longitudinal and lateral road surface slopes, which, together with in-vehicle IMU and GNSS/INS, will enable a more accurate decomposition of linear and gravitation-caused acceleration.…”

Research on Front-Vehicle taillight recognition based on image processing and matching

“…The improved EIOU-yolov5 algorithm can better assist clinical diagnosis. I [21] proposed an improved automobile taillight recognition algorithm based on YOLOv5s, and the final mAP reached 95.3%, effectively identifying the automobile taillight.…”