一种芯片表面缺陷检测方法及装置、系统、存储介质

By combining detection methods using bright-field and dark-field images and employing confidence scores for evaluation, the problem of balancing high resolution and high throughput in biochip surface defect detection has been solved, achieving higher precision defect detection.

CN122115461BActive Publication Date: 2026-07-17ANHUI SHUZIXING INTELLIGENT TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
ANHUI SHUZIXING INTELLIGENT TECH CO LTD
Filing Date
2026-04-30
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

In existing technologies, it is difficult to achieve both high resolution and high throughput for the detection of surface defects on biochips, resulting in insufficient detection accuracy.

Method used

A method combining bright-field and dark-field images is adopted. An initial image group is obtained through image registration, the classification and prediction distribution of defect primitives are determined, and a confidence score is constructed using the first feature information and the second feature information to determine whether the defect primitives need to be resampled to improve the detection accuracy.

Benefits of technology

By combining bright and dark field images and using confidence scores for evaluation, the accuracy and detail representation of surface defects in biochips are improved, ensuring the accuracy of the detection results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122115461B_ABST
    Figure CN122115461B_ABST
Patent Text Reader

Abstract

本发明涉及生物芯片缺陷检测技术领域,公开了一种芯片表面缺陷检测方法及装置、系统、存储介质,本发明通过获取生物芯片的明暗场图像,提高表面缺陷的可辨识度。然后,基于缺陷图元的分类预测分布,构建置信度分数,通过第一特征信息与第二特征信息,分别从同一缺陷图元在不同图像之间的“一致性差异”与同一缺陷预测类型的不同图元之间的“个体性差异”的角度,丰富置信度分数的评估标准,进一步精确置信度分数。当该分数小于第一阈值时,认定该缺陷图元为低置信度,并对其对应位置的图像进行重新采样,进一步提高了图像精度和图像细节表达,使缺陷检测过程和得到的结果更准确,从而提高了生物芯片表面缺陷检测的精准度。
Need to check novelty before this filing date? Find Prior Art