A food packaging sealing quality detection method and system

By using machine vision technology to perform coarse and fine positioning of food packaging seals, combined with detection and segmentation models, the problem of the inability to detect minute sealing defects in existing technologies has been solved, improving detection accuracy and reducing costs.

CN122409684APending Publication Date: 2026-07-17SIGMA SQUARES (BEIJING) TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SIGMA SQUARES (BEIJING) TECH CO LTD
Filing Date
2026-04-13
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing food packaging sealing inspection devices cannot effectively detect minute and subtle defects, such as wrinkles, broken edges, and oil stains, and the inspection cost is high, failing to meet food safety and quality requirements.

Method used

Using machine vision technology, coarse and fine localization are performed by acquiring food packaging images. Combined with pre-built detection and segmentation models, defects in the sealing area are detected, including defect segmentation in the coarse localization sealing area and defect verification in the fine localization sealing area, and valid defects are screened out.

Benefits of technology

It improves the accuracy of food packaging sealing quality inspection, reduces the rate of missed detection, and can effectively detect minor sealing defects, ensuring food safety and quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122409684A_ABST
    Figure CN122409684A_ABST
Patent Text Reader

Abstract

本申请涉及机器视觉技术领域,具体是一种食品包装封口质量检测方法及系统,本申请先获取食品包装图像;对食品包装图像的封口区域进行粗定位,得到粗定位封口区域;基于预先构建的分割模型对粗定位封口区域进行瑕疵检测,得到粗定位封口区域的瑕疵分割结果;基于粗定位封口区域与食品包装图像的相对位置判断封口的方向,并基于封口的方向在食品包装图像中进行精细定位,得到精定位封口区域;对位于精定位封口区域内的瑕疵分割结果进行校验,得到有效瑕疵。本申请通过粗定位后精确定位封口区域解决封口区域外的瑕疵误判问题,分割模型可以检测多种缺陷,降低了目前市面上的食品包装封口质量检测方法的漏检率,提高了食品包装封口质量的检测准确率。
Need to check novelty before this filing date? Find Prior Art