Screw quality monitoring method and system based on cross-process visual traceability

By assigning a unique identifier to the screw and utilizing deep embedding and sequence anomaly detection models, a visual feature profile of the screw is constructed, solving the problems of coarse granularity in screw quality traceability and isolated detection data, and realizing rapid and accurate defect root cause analysis and preventive quality control.

CN122415533APending Publication Date: 2026-07-17ZHEJIANG ZHUOYOU GENERAL MASCH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
ZHEJIANG ZHUOYOU GENERAL MASCH CO LTD
Filing Date
2026-04-23
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies for screw quality traceability are coarse-grained, and visual inspection data is isolated, making it difficult to achieve rapid and accurate defect root cause analysis and traceability.

Method used

By adopting a cross-process visual traceability approach, each screw is assigned a machine-readable unique identifier in the screw production process. Combined with a deep embedding model and a sequence anomaly detection model, a visual feature profile is constructed that runs through the production cycle, enabling continuous monitoring of screw quality and location of defect roots.

Benefits of technology

It enables refined quality traceability, allowing for quick and accurate identification of the root cause of defects, reducing quality costs and material waste, and supporting rapid response and preventative quality maintenance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122415533A_ABST
    Figure CN122415533A_ABST
Patent Text Reader

Abstract

本发明公开了一种基于跨工序视觉溯源的螺杆品质监控方法及系统,属于工业产品质量监控技术领域。该方法包括:在螺杆生产的首个工序中,为所述螺杆赋予唯一身份标识;在所述螺杆生产过程中的多个预设工序节点,分别获取与所述唯一身份标识关联的螺杆在制品图像;采用预先训练的深度嵌入模型,将所述多个预设工序节点获取的螺杆在制品图像,分别编码为对应的特征向量,以建立贯穿多个工序的、所述螺杆的特征向量序列;在接收到针对所述螺杆的缺陷溯源指令时,调用预先训练的序列异常检测模型,对所述特征向量序列进行分析,以确定引入缺陷的目标工序。本发明能够快速、自动化地定位缺陷根源工序,为工艺优化提供数据支撑。
Need to check novelty before this filing date? Find Prior Art