基于Mask-RCNN的金属缺陷识别系统设计

Design of Metal Defect Recognition System Based on Mask-RCNN

在传统的工业生产中,特别是铸造或焊接的生产过程,对金属件的缺陷识别是非常重要的。传统的人工缺陷识别程序通常耗时且容易出错。然而,通过引入智能化缺陷识别系统,可以提高检测效率,满足生产高质量产品的需求。文中提出了一种基于Mask-RCNN模型的X射线无损检测缺陷识别系统。该系统使用客户-服务器模式,由客户端上传原始图片,服务器调用Mask-RCNN模型将原始图片进行缺陷识别,返回识别缺陷后的图片给客户端,在客户端上通过掩模透明化功能进行缺陷识别结果,并分析比对。该系统以Mask-RCNN模型进行缺陷识别,完成了缺陷识别工作,解决了传统人工缺陷识别程序易出错、耗时多的问题,并提升了铸件或焊接件的缺陷检识别效率和可靠性,解放了执行缺陷识别工作质检人员的生产力。

In traditional industrial production,especially in casting or welding production processes,the identification of defects in metal parts is very important.Traditional manual defect identification procedures are often time-consuming and error-prone.However,by introducing an intelligent defect recognition system,inspection efficiency can be improved to meet the demand for producing high-quality products.This paper proposes an X-ray non-destructive inspection defect identification system based on the Mask-RCNN model.The system uses the client-server model.The client uploads the original image.The server calls the Mask-RCNN model to identify defects in the original image and returns the image after identifying the defects to the client.The mask transparency function is used on the client.Analysis and comparison of defect identification results.The system uses the Mask-RCNN model to identify defects,completes the defect identification work,solves the error-prone and time-consuming problems of traditional manual defect identification procedures,and improves the efficiency and reliability of defect detection and identification of castings or weldments,liberating the productivity of quality inspection personnel who perform defect identification work.