arXiv 2501.125962025-01cs.CVstat.APstat.OT

适配 OpenAI 的 CLIP 模型用于制造业质量控制中的少样本图像检测：一项包含多个应用示例的说明性案例研究

Adapting OpenAI's CLIP Model for Few-Shot Image Inspection in Manufacturing Quality Control: An Expository Case Study with Multiple Application Examples

Farmer School of Business, Miami University（迈阿密大学法默商学院）；College of Arts and Sciences, University of Dayton（Dayton大学文理学院）；Department of Mechanical Engineering, Politecnico di Milano（米兰理工学院机械工程系）；Mathematics & Statistics, Helmut Schmidt University（海姆-施密特大学数学与统计学系）；College of Engineering, University of Georgia（佐治亚大学工程学院）；Amsterdam Business School, University of Amsterdam（阿姆斯特丹大学阿姆斯特丹商学院）

Fadel M. Megahed, Ying-Ju Chen, Bianca Maria Colosimo, Marco Luigi Giuseppe Grasso, L. Allison Jones-Farmer, Sven Knoth, Hongyue Sun, Inez Zwetsloot

备注：: 36 pages, 13 figures

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出将 OpenAI 的 CLIP 模型适配于制造业质量检测的少样本学习，通过五个案例评估表明该方法在单组件和纹理应用中实现高准确率，并提供实用框架作为有效基线。

摘要：

这篇说明性论文介绍了一种简化的制造业图像质量检测方法，该方法使用适配于少样本学习的 OpenAI 的 CLIP（Contrastive Language-Image Pretraining，对比语言-图像预训练）模型。尽管 CLIP 在通用计算机视觉任务中展现了出色的能力，但由于其训练数据与工业应用之间存在领域差距，将其直接应用于制造检测面临挑战。我们通过五个案例研究评估了 CLIP 的有效性：金属锅表面检测、3D 打印挤出轮廓分析、随机纹理表面评估、汽车装配检测和微观结构图像分类。结果表明，对于单组件和基于纹理的应用，CLIP 能够在使用相对较小的学习集（每类 50-100 个示例）的情况下实现高分类准确率。然而，在复杂的多组件场景下性能会下降。我们提供了一个实用的实现框架，使质量工程师能够在寻求更复杂的解决方案之前，快速评估 CLIP 对其特定应用的适用性。这项工作确立了基于 CLIP 的少样本学习作为一种有效的基线方法，它在实现的简单性与稳健的性能之间取得了平衡，并在多个制造质量控制应用中得到了证明。

英文摘要：

This expository paper introduces a simplified approach to image-based quality inspection in manufacturing using OpenAI's CLIP (Contrastive Language-Image Pretraining) model adapted for few-shot learning. While CLIP has demonstrated impressive capabilities in general computer vision tasks, its direct application to manufacturing inspection presents challenges due to the domain gap between its training data and industrial applications. We evaluate CLIP's effectiveness through five case studies: metallic pan surface inspection, 3D printing extrusion profile analysis, stochastic textured surface evaluation, automotive assembly inspection, and microstructure image classification. Our results show that CLIP can achieve high classification accuracy with relatively small learning sets (50-100 examples per class) for single-component and texture-based applications. However, the performance degrades with complex multi-component scenes. We provide a practical implementation framework that enables quality engineers to quickly assess CLIP's suitability for their specific applications before pursuing more complex solutions. This work establishes CLIP-based few-shot learning as an effective baseline approach that balances implementation simplicity with robust performance, demonstrated in several manufacturing quality control applications.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。