The intelligent black smoke vehicle recognition method based on surveillance video can effectively save man-power and material resources,and has broad application prospects.However,the black smoke emitted by vehicles is semi transparent and difficult to distinguish from the asphalt pavement in the background.As the vehicle moves,the black smoke will generate smoke plumes and have unstable shapes,ultimately resulting in low accuracy and recall of black smoke recognition.Firstly,use the YOLOv5s-MobileNetv3 model to locate the vehicle and capture the smoke exhaust area to reduce the amount of data for subsequent processing;secondly,the K-Means algorithm is used to cluster the black smoke at the rear of the vehicle to obtain a universal aspect ratio,and the exhaust area at the rear of the vehicle is ex-tracted based on the aspect ratio;finally,a dual teacher combined distillation black smoke recognition method is proposed for vehicle exhaust black smoke recognition.Training and testing are conducted on 62 surveillance videos of a certain highway section,including black smoke vehicles.The target vehicle detection speed is 76 fps,with an accuracy of 94.70%.The recall rate is 97.50%,the accuracy rate of black smoke feature recognition is 93.43%,and the false alarm rate was 6.52%.The use of lightweight networks for vehicle localization reduces algorithm complexity and ensures real-time performance of the method;the dual teacher joint distillation network model proposed in this article has significant advantages in recognition time while ensuring high accuracy.
vehicle detectionYOLOv5sblack smoke exhaust recognitionknowledge distillationteacher student model
维普
万方数据
