TaylorMade VDD: Domain-adaptive Visual Defect Detector for High-mix Low-volume Production of Non-convex Cylindrical Metal Objects

TL;DR

This paper introduces a domain-adaptive visual defect detection framework using neural architecture search, specifically designed for high-mix low-volume production of non-convex cylindrical metal objects, improving defect detection accuracy across domain shifts.

Contribution

The paper presents a novel VDD framework that automatically adapts to new domains via NAS, addressing domain shift challenges in defect detection for complex metal objects.

Findings

Achieved higher burr detection accuracy across different domains.

Demonstrated effectiveness on non-convex cylindrical metal objects.

Outperformed baseline methods in domain adaptation scenarios.

Abstract

Visual defect detection (VDD) for high-mix low-volume production of non-convex metal objects, such as high-pressure cylindrical piping joint parts (VDD-HPPPs), is challenging because subtle difference in domain (e.g., metal objects, imaging device, viewpoints, lighting) significantly affects the specular reflection characteristics of individual metal object types. In this paper, we address this issue by introducing a tailor-made VDD framework that can be automatically adapted to a new domain. Specifically, we formulate this adaptation task as the problem of network architecture search (NAS) on a deep object-detection network, in which the network architecture is searched via reinforcement learning. We demonstrate the effectiveness of the proposed framework using the VDD-HPPPs task as a factory case study. Experimental results show that the proposed method achieved higher burr detection accuracy compared with the baseline method for data with different training/test domains for the non-convex HPPPs, which are particularly affected by domain shifts.