Defect Detection of Forming Mesh Based on YOLOV8N Model

A method for detecting defects in forming mesh based on the YOLOV8N model is proposed. Firstly, an illumination system is designed, and then image data with defects is collected to construct an image dataset. The constructed image dataset is annotated and augmented to build a corresponding image sample set. The YOLOV8N model is trained using the sample set to obtain a defect recognition and localization model. The original image to be recognized is input into the defect recognition and localization model, which then outputs the corresponding defect recognition and localization results. Experimental results show that this method provides accurate localization and fast speed.