柔性玻璃盖板表面缺陷检测方法及其装置

By binarizing images of flexible glass covers and fusing clustering algorithms with recursive algorithms, surface defects of flexible glass covers can be automatically detected, solving the problems of high cost and low efficiency of manual inspection and achieving efficient and stable defect identification.

CN116703804BActive Publication Date: 2026-07-17TOMI CHENGDU APPLIED TECH RES INST CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
TOMI CHENGDU APPLIED TECH RES INST CO LTD
Filing Date
2022-02-24
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

In the current technology, the detection of surface defects of flexible glass covers for smartphones mainly relies on manual visual inspection, which is costly, inefficient, difficult to detect and has unstable results. In particular, it is harmful to the human eye in strong light environments and is difficult to detect tiny defects.

Method used

By acquiring images of flexible glass covers and performing binarization processing, the target connected components and center points are determined. Then, using a chain-like connection method combining clustering and recursive algorithms, dense point defect areas are automatically detected.

Benefits of technology

It enables automated detection of surface defects in flexible glass covers, reduces labor costs, improves detection efficiency and accuracy, reduces false detections, and meets the requirements for real-time online detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116703804B_ABST
    Figure CN116703804B_ABST
Patent Text Reader

Abstract

本申请提出一种柔性玻璃盖板表面缺陷检测方法及其装置,涉及柔性玻璃检测技术领域,可应用于智能手机柔性玻璃盖板表面缺陷的检测场景中。具体实现方案为:获取柔性玻璃盖板的图像;对图像进行二值化处理,获得二值化图像;确定二值化图像中符合缺陷条件的至少一个目标连通域,并确定每个目标连通域中心点;根据密集点的链式连接方式,采用聚类算法融合递归算法的搜索方式对每个目标连通域中心点进行运算,以从至少一个目标连通域中确定出密集点缺陷区域。本申请实现了对于柔性玻璃盖板表面缺陷的自动化检测,提高了对于柔性玻璃盖板表面缺陷的检测效率,提高了柔性玻璃盖板表面缺陷识别的稳定性与精准性。
Need to check novelty before this filing date? Find Prior Art