Enhancing Glass Defect Detection with Diffusion Models: Addressing Imbalanced Datasets in Manufacturing Quality Control

2025-05-06

AI Summary

Industrial glass defect detection suffers from severe class imbalance due to the scarcity of authentic defect samples, significantly limiting deep learning model performance. To address this, we propose the first application of Denoising Diffusion Probabilistic Models (DDPMs) for industrial glass defect data augmentation—generating physically plausible and high-fidelity synthetic defect images without manual annotation. We integrate this augmented dataset with multiple CNN backbones—including ResNet50V2, EfficientNetB0, and MobileNetV2—and employ a precision-recall co-optimization strategy to mitigate bias under limited-sample conditions. Experimental results demonstrate that ResNet50V2 achieves a substantial accuracy improvement from 78% to 93%, while defect recall increases markedly without compromising precision (maintained at 100%). These outcomes validate the proposed method’s effectiveness, robustness, and generalizability in real-world production-line quality inspection.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Search and Optimization: Learning to Search

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Visual defect detection in industrial glass manufacturing remains a critical challenge due to the low frequency of defective products, leading to imbalanced datasets that limit the performance of deep learning models and computer vision systems. This paper presents a novel approach using Denoising Diffusion Probabilistic Models (DDPMs) to generate synthetic defective glass product images for data augmentation, effectively addressing class imbalance issues in manufacturing quality control and automated visual inspection. The methodology significantly enhances image classification performance of standard CNN architectures (ResNet50V2, EfficientNetB0, and MobileNetV2) in detecting anomalies by increasing the minority class representation. Experimental results demonstrate substantial improvements in key machine learning metrics, particularly in recall for defective samples across all tested deep neural network architectures while maintaining perfect precision. The most dramatic improvement was observed in ResNet50V2's overall classification accuracy, which increased from 78 percent to 93 percent when trained with the augmented data. This work provides a scalable, cost-effective approach to enhancing automated defect detection in glass manufacturing that can potentially be extended to other industrial quality assurance systems and industries with similar class imbalance challenges.

Problem

• Addressing imbalanced datasets in glass defect detection
• Enhancing deep learning performance with synthetic defect images
• Improving automated quality control in manufacturing systems

Innovation

• Using DDPMs to generate synthetic defective images
• Enhancing CNN performance with augmented data
• Improving recall for defective samples significantly