Feature marker detection method and device for iron and steel industry

The invention provides a feature marker detection method and device for the iron and steel industry, and can be applied to the technical field of iron and steel metallurgy. According to the specific implementation scheme, the method comprises the steps of obtaining feature marker image data in a preset scene, performing preprocessing operation on the feature marker image data, and constructing a feature marker data set of the iron and steel industry; constructing an improved YOLOv5 neural network model, and training the improved YOLOv5 neural network model based on the feature marker data set and a K-Means + + clustering algorithm until the improved YOLOv5 neural network model meets a preset model evaluation index, and obtaining a trained feature marker detection model; and deploying the feature marker detection model in the preset scene, detecting a to-be-detected feature marker image in the preset scene based on the feature marker detection model, and determining a feature marker type corresponding to the t