SIAD: Self-supervised Image Anomaly Detection System

TL;DR

SIAD introduces a self-supervised system for continuous, online visual inspection in manufacturing, reducing human intervention and supporting the entire product lifecycle with automatic annotation and adaptable algorithms.

Contribution

The paper presents SsaA, a novel self-supervised annotation system enabling long-term, online visual inspection in manufacturing without human-in-the-loop, integrating unsupervised and supervised learning.

Findings

Successfully applied in real industrial scenarios

Supports entire manufacturing lifecycle

Automates annotation process

Abstract

Recent trends in AIGC effectively boosted the application of visual inspection. However, most of the available systems work in a human-in-the-loop manner and can not provide long-term support to the online application. To make a step forward, this paper outlines an automatic annotation system called SsaA, working in a self-supervised learning manner, for continuously making the online visual inspection in the manufacturing automation scenarios. Benefit from the self-supervised learning, SsaA is effective to establish a visual inspection application for the whole life-cycle of manufacturing. In the early stage, with only the anomaly-free data, the unsupervised algorithms are adopted to process the pretext task and generate coarse labels for the following data. Then supervised algorithms are trained for the downstream task. With user-friendly web-based interfaces, SsaA is very convenient to integrate and deploy both of the unsupervised and supervised algorithms. So far, the SsaA system has been adopted for some real-life industrial applications.