Real-time Traffic Signs Detection Based on YOLO Network Model
Wenkao Yang, Wei Zhang
Abstract
Recently, real-time traffic sign detection has been widely applied in autonomous and assisted car driving. At the same time, the research on traffic signs detection based on YOLO (You Only Look Once) has attracted intensive attention. However, the accuracy and precision of small targets detection based on YOLO need to be further improved. We trained and tested the latest YOLOv4 and YOLOv3 on the same data set to comparatively study the detection results. The data set consists of 4000 Chinese traffic signs, which were manually labeled by ourselves. Comparing the detection results, the detection accuracy of the V4 is significantly higher than that of the V3.
Topics & Concepts
Computer scienceObject detectionSet (abstract data type)Artificial intelligenceData setReal-time computingComputer visionData miningPattern recognition (psychology)Programming languageAdvanced Neural Network ApplicationsVideo Surveillance and Tracking MethodsVehicle License Plate Recognition