Recognition Algorithm for Coke Oven Smoke and Fire Based on Improved YOLOv5s
For the requirements of all-weather environmental monitoring of smoke and fire emissions in coke plants,a coke oven smoke and fire recognition algorithm based on improved YOLOv5s is proposed;the algorithm uses YOLOv5s as the base network and adds the attention mechanism module of convolutional block attention module(CBAM)to the backbone network,it makes the net-work pay more attention to the important features and improve the detection accuracy of targets;a new Sigmoid weighted liner unit(FReLU)activation function replaces the funnel rectified linear unit(SiLU)activation function to improve the sensitivity of the acti-vation space and the smoke and fire image vision task;on the basis of smoke and fire sample labels in the self-built dataset,the light labels are added to solve the interference of strong lights on flame recognition,and the smoke and fire detection problem of day and night scenes is achieved by the shunting training and detection;Through the comparison experiments on the self-built dataset and re-placing the activation function,the experimental results for the joint CBAM module show that the mAP value of smoke and fire detec-tion in the day scene is improved by 6.7％than that of the original YOLOv5s model,and the mAP value of smoke and fire recognition in night scenes reaches 97.4％.
smoke and fire recognitionYOLOv5sattention mechanismactivation functiontarget detection
万方数据
维普
