Screening method of object surface morphology based on artificial neural network

Through the surface pattern screening method of object based on artificial neural networks, multiple sub-neural network systems are used to form a complete neural network system, which solves the problems of low efficiency and easy misjudgment of surface defects in structural objects in the prior art, and achieves efficient and accurate object screening.

CN112683924BActive Publication Date: 2025-03-18SHENXUN COMP KUNSHAN +1
1 Cites 0 Cited by

Patent Information

Application Number
CN201910987177.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-10-17
Publication Date
2025-03-18
Estimated Expiration
2039-10-17

AI Technical Summary

Technical Problem

The prior art is inefficient and prone to misjudgment when detecting surface defects of structural objects, resulting in the inability to control the yield of structural objects.

Method used

The surface pattern screening method of object based on artificial neural network is adopted. By receiving object images, multiple sub-neural network systems are used to form a complete neural network system to identify and classify surface pattern to accurately screen objects.

Benefits of technology

It realizes rapid and accurate classification of object images, reduces over-release rate, and improves the detection efficiency and accuracy of structural objects.

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

Abstract

A method for screening object surface morphology based on an artificial neural network includes: receiving at least one object image; using a first prediction model to perform surface morphology recognition on each object image to classify it into a first normal group and a first abnormal group; and using a second prediction model to perform surface morphology recognition on each output image in the first normal group to classify it into a second normal group and a second abnormal group. The method for screening object surface morphology based on an artificial neural network of the present invention continuously performs surface morphology recognition on object images through multiple neural networks with different training conditions connected in series to accurately and quickly classify object images, thereby efficiently screening objects corresponding to the object images based on the classification results of the object images, thereby obtaining a lower over-release rate.
Need to check novelty before this filing date? Find Prior Art