Industrial image anomaly detection method and system based on feature-level anomaly synthesis and reconstruction

By extracting and adapting multi-scale features of industrial images, adding noise to simulate anomalies and reconstructing normal regions, the problems of high computational cost and unrealistic synthesized anomalies in existing technologies are solved, achieving high-accuracy industrial image anomaly detection and improving production efficiency and product quality.

CN120598893BActive Publication Date: 2026-04-28HANGZHOU DIANZI UNIV
1 Cites 0 Cited by

Patent Information

Application Number
CN202510695060.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-05-28
Publication Date
2026-04-28
Estimated Expiration
2045-05-28

AI Technical Summary

Technical Problem

Existing industrial image anomaly detection methods are computationally expensive, produce unrealistic anomalies, and are difficult to reconstruct anomaly regions effectively, resulting in low detection accuracy.

Method used

By acquiring visible light images of products from industrial production lines, multi-scale local features are extracted, the size is standardized and the features are adapted, noise is added to simulate anomalies, a feature reconstruction network is used to reconstruct normal regions, feature map difference loss is calculated to train the model, and segmentation maps are generated for detection.

Benefits of technology

It improves the accuracy and speed of industrial image anomaly detection, reduces computing costs, increases production efficiency and product quality, and promotes the intelligentization of industrial systems.

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

Abstract

The application discloses an industrial image anomaly detection method and system based on feature-level abnormal synthesis and reconstruction, and the method is as follows: S1, acquiring an image, and extracting multi-scale local features through a pre-trained network; S2, adjusting the features of different scales to a unified size, connecting the features of the unified size, and obtaining a feature map; S3, transferring the feature map of S2 to a target domain through feature adaptation, and obtaining a new feature map; S4, adding noise to the feature map obtained in S3, and obtaining a feature map simulating an anomaly; S5, inputting the feature map obtained in S4 into a feature reconstruction network, and reconstructing the features into features without anomalies; S6, calculating a loss through the difference between the feature map before adding noise and the reconstructed feature map, training the feature reconstruction network according to the loss value; and S7, reconstructing the features by using the trained network, and obtaining a segmentation map according to the difference before and after the feature reconstruction. The application solves the problems of high calculation cost, unreal synthesized anomaly and reconstructed anomaly area of the existing anomaly detection method.
Need to check novelty before this filing date? Find Prior Art