Training method of chip fuzzy defect detection model, defect detection method and equipment

By training the chip fuzzy defect detection model, using feature extractors and self-supervised classifiers, the problem of difficult and cost of chip defect detection in the existing technology is solved, and efficient and accurate fuzzy defect detection is achieved.

CN119722589AActive Publication Date: 2025-03-28CASIC DEFENSE TECH RES & TEST CENT
5 Cites 0 Cited by

Patent Information

Application Number
CN202411715050.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-27
Publication Date
2025-03-28
Estimated Expiration
2044-11-27

AI Technical Summary

Technical Problem

Existing chip defect detection methods are difficult to accurately detect fuzzy defects of chips in complex industrial production environments, and rely on high sensitivity detection devices and a large number of human resources, resulting in high detection costs and low efficiency.

Method used

A training method for chip fuzzy defect detection model is proposed. By acquiring the image data set and the initial detection model, using feature extractor, self-supervised classifier and detector for training, a detection model that can accurately identify defects under fuzzy conditions is generated.

Benefits of technology

It realizes efficient and accurate detection of fuzzy defects of the chip without relying on high-sensitivity detection devices and human resources, reducing detection costs and improving detection efficiency.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a training method of a chip fuzzy defect detection model, a defect detection method and equipment. The training method comprises the following steps: acquiring an image data set and an initial chip fuzzy defect detection model; performing feature extraction on the image data by using a feature extractor to obtain image features; inputting the image features into an initial self-supervised classifier, and determining a self-supervised loss function corresponding to the initial self-supervised classifier; inputting the image features into an initial detector, processing the image features by the initial detector, and outputting predicted defect data; determining a detection loss function according to the real defect data and the predicted defect data; adding the self-supervision loss function and the detection loss function to obtain a target loss function; and training the initial chip fuzzy defect detection model based on the target loss function until the number of training times of the initial chip fuzzy defect detection model reaches a preset number of times, thereby obtaining the chip fuzzy defect detection model.
Need to check novelty before this filing date? Find Prior Art