Improving Cumulative Coal Recognition in YOLOv5s Network

In order to solve the problem of not being able to effectively detect the coal accumulation and clean the coal accumulation area in the traditional mechanical cleaning of coal grates, this paper proposes an improved coal accumulation detection method for coal grates based on YOLOv5s. Firstly, the FasterNet structure is used to replace the CSPDarknet structure to reduce the number of floating-point operations of the network and solve the problem of large arithmetic requirements; secondly, the spatial pyramid module is improved, and the SPPF structure is replaced by the SPPFCSPC structure to improve the detection capability of the network for targets of different sizes; finally, the SIOU loss function is used to accelerate the convergence of the model, and to optimize the training process of the model. The experimental results show that the improved models mAP and Precision are improved by 1.3% and 1.1% respectively compared to the original model.