Small sample image anomaly detection method based on multi-scale attention and efficient convolution

By constructing a small sample image anomaly detection method of ResNet network and multi-scale attention mechanism, the problem of failing to fully utilize early features in the prior art is solved, and efficient anomaly detection under small sample conditions is achieved.

CN120298770APending Publication Date: 2025-07-11CHANGZHOU UNIV
0 Cites 2 Cited by

Patent Information

Application Number
CN202510357469.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-03-25
Publication Date
2025-07-11

AI Technical Summary

Technical Problem

The existing small sample image abnormality detection methods fail to make full use of features extracted in the early stage of encoding, resulting in insufficient accuracy of abnormality detection.

Method used

A small sample image anomaly detection method based on multi-scale attention and efficient convolution is adopted. By constructing a ResNet network encoder, combining MSCAM, EUCB and LGAG modules, semantic information at different levels of the image is integrated, and feature mapping is enhanced by using a multi-scale convolution attention mechanism, and anomaly detection is performed by combining a probability judgment classifier of multivariate Gaussian distribution.

Benefits of technology

When the number of training samples is limited and the test sample label is unknown, the model's understanding of normal sample distribution is improved, the sensitivity to abnormal points and detection accuracy is enhanced, and the semantic information at different levels of the image is effectively integrated, which improves the accuracy of abnormal detection.

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

Abstract

The invention relates to the technical field of image processing, in particular to a small sample image anomaly detection method based on multi-scale attention and efficient convolution, which comprises the following steps: acquiring small sample image data; the method comprises the following steps: constructing a ResNet network encoder, and constructing decoders of an MSCAM module, an EUCB module and an LGAG module; image features are extracted through an encoder, layer output feature mapping from low levels to high levels is constructed, and different levels of feature information are learned to integrate different levels of semantic information of the image; enhancing feature mapping in the decoder by using a multi-scale convolution attention mechanism; a complex spatial relationship and local attention are integrated by using channel, space and grouping gating attention mechanisms; and output vectors of different levels of the network are spliced. The problem that the accuracy of anomaly detection needs to be further improved due to the fact that features extracted in the early stage of coding are not fully utilized in an existing method is solved.
Need to check novelty before this filing date? Find Prior Art