Research on bubble defect detection algorithm of epoxy encapsulation products based on YOLOv8

This paper mainly focuses on the research of the detection of internal bubble defects in epoxy encapsulation products. Firstly, the characteristics of bubble defects in CT images of epoxy encapsulation products are analyzed. Based on these characteristics, the segmentation model from the YOLOv8 is selected to detect and segment the bubble defects in the CT images. In addition, in order to further improve the segmentation effect, this paper also combines image processing algorithms to more accurately separate the bubble defects from the original images. Finally, through the application of visualization software, the display and analysis of the bubble defect detection results become more intuitive and efficient. This provides a powerful technical support for quality control in industrial production process and helps improve production efficiency and product quality. The experimental results show that the segmentation model from the YOLOv8 can not only effectively identify and segment bubble defects from the CT images of epoxy encapsulation products, but also achieves high segmentation accuracy, particularly suited for batch processing tasks.