Xinjian Jia, Tongcai Wang, Yizhe Yang, Xiaodong Liu, Xin Li, Bingshan Liu, Gong Wang
{"title":"基于多图像融合的缺陷检测方法,用于实时监控陶瓷增材制造中的重涂。","authors":"Xinjian Jia, Tongcai Wang, Yizhe Yang, Xiaodong Liu, Xin Li, Bingshan Liu, Gong Wang","doi":"10.1089/3dp.2023.0285","DOIUrl":null,"url":null,"abstract":"<p><p>Vat photopolymerization is characterized by its high precision and efficiency, making it a highly promising technique in ceramic additive manufacturing. However, the process faces a significant challenge in the form of recoating defects, necessitating real-time monitoring to maintain process stability. This article presents a defect detection method that leverages multi-image fusion and deep learning for identifying recoating defects in ceramic additive manufacturing. In the image fusion process, multiple single-channel recoating images captured by monitoring camera positioned near the photopolymerization equipment are merged with curing area mask image to create a three-channel color image. The recoating images suffer from perspective distortion due to their side view. To facilitate fusion with the curing area image, image rectification technique is applied to correct the perspective distortion, transforming the side view recoating images into a top-down view. Subsequently, the fused images are processed using a channel-wise YOLO (You Only Look Once, CW-YOLO) method to extract features, enabling the distinction of various types of defects. When compared with other deep learning models, CW-YOLO achieves higher detection accuracy while maintaining a detection rate of 103.58fps, meeting the requirements for real-time detection. Furthermore, the paper introduces the F1 score as a comprehensive evaluation metric, capturing both detection accuracy and recall rate. The results show that the F1 score is enhanced by approximately 10% after image fusion, demonstrating that the proposed method can significantly improve defect detection, particularly in cases involving difficult-to-distinguish defects like material shortages and scratches.</p>","PeriodicalId":54341,"journal":{"name":"3D Printing and Additive Manufacturing","volume":"12 1","pages":"11-22"},"PeriodicalIF":2.3000,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11937757/pdf/","citationCount":"0","resultStr":"{\"title\":\"Multi-Image Fusion-Based Defect Detection Method for Real-Time Monitoring of Recoating in Ceramic Additive Manufacturing.\",\"authors\":\"Xinjian Jia, Tongcai Wang, Yizhe Yang, Xiaodong Liu, Xin Li, Bingshan Liu, Gong Wang\",\"doi\":\"10.1089/3dp.2023.0285\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Vat photopolymerization is characterized by its high precision and efficiency, making it a highly promising technique in ceramic additive manufacturing. However, the process faces a significant challenge in the form of recoating defects, necessitating real-time monitoring to maintain process stability. This article presents a defect detection method that leverages multi-image fusion and deep learning for identifying recoating defects in ceramic additive manufacturing. In the image fusion process, multiple single-channel recoating images captured by monitoring camera positioned near the photopolymerization equipment are merged with curing area mask image to create a three-channel color image. The recoating images suffer from perspective distortion due to their side view. To facilitate fusion with the curing area image, image rectification technique is applied to correct the perspective distortion, transforming the side view recoating images into a top-down view. Subsequently, the fused images are processed using a channel-wise YOLO (You Only Look Once, CW-YOLO) method to extract features, enabling the distinction of various types of defects. When compared with other deep learning models, CW-YOLO achieves higher detection accuracy while maintaining a detection rate of 103.58fps, meeting the requirements for real-time detection. Furthermore, the paper introduces the F1 score as a comprehensive evaluation metric, capturing both detection accuracy and recall rate. The results show that the F1 score is enhanced by approximately 10% after image fusion, demonstrating that the proposed method can significantly improve defect detection, particularly in cases involving difficult-to-distinguish defects like material shortages and scratches.</p>\",\"PeriodicalId\":54341,\"journal\":{\"name\":\"3D Printing and Additive Manufacturing\",\"volume\":\"12 1\",\"pages\":\"11-22\"},\"PeriodicalIF\":2.3000,\"publicationDate\":\"2025-02-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11937757/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"3D Printing and Additive Manufacturing\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1089/3dp.2023.0285\",\"RegionNum\":4,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/2/1 0:00:00\",\"PubModel\":\"eCollection\",\"JCR\":\"Q3\",\"JCRName\":\"ENGINEERING, MANUFACTURING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"3D Printing and Additive Manufacturing","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1089/3dp.2023.0285","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/2/1 0:00:00","PubModel":"eCollection","JCR":"Q3","JCRName":"ENGINEERING, MANUFACTURING","Score":null,"Total":0}
引用次数: 0
摘要
还原光聚合具有精度高、效率高等特点,在陶瓷增材制造中具有广阔的应用前景。然而,该工艺面临着重涂缺陷的重大挑战,需要实时监控以保持工艺稳定性。本文提出了一种利用多图像融合和深度学习来识别陶瓷增材制造中重涂缺陷的缺陷检测方法。在图像融合过程中,将位于光聚合设备附近的监控摄像机捕获的多个单通道重涂图像与固化区域掩模图像合并,形成三通道彩色图像。重绘图像由于其侧面视图而遭受透视失真。为了便于与固化区域图像融合,采用图像校正技术对透视畸变进行校正,将侧视图重绘图像转换为自上而下的视图。随后,采用逐通道YOLO (You Only Look Once, CW-YOLO)方法对融合后的图像进行特征提取,实现了不同类型缺陷的区分。与其他深度学习模型相比,CW-YOLO在保持103.58fps的检测率的同时,实现了更高的检测精度,满足了实时检测的要求。此外,本文引入F1分数作为综合评价指标,同时捕获检测准确率和召回率。结果表明,图像融合后的F1分数提高了约10%,表明所提出的方法可以显著改善缺陷检测,特别是在材料短缺和划痕等难以区分的缺陷情况下。
Multi-Image Fusion-Based Defect Detection Method for Real-Time Monitoring of Recoating in Ceramic Additive Manufacturing.
Vat photopolymerization is characterized by its high precision and efficiency, making it a highly promising technique in ceramic additive manufacturing. However, the process faces a significant challenge in the form of recoating defects, necessitating real-time monitoring to maintain process stability. This article presents a defect detection method that leverages multi-image fusion and deep learning for identifying recoating defects in ceramic additive manufacturing. In the image fusion process, multiple single-channel recoating images captured by monitoring camera positioned near the photopolymerization equipment are merged with curing area mask image to create a three-channel color image. The recoating images suffer from perspective distortion due to their side view. To facilitate fusion with the curing area image, image rectification technique is applied to correct the perspective distortion, transforming the side view recoating images into a top-down view. Subsequently, the fused images are processed using a channel-wise YOLO (You Only Look Once, CW-YOLO) method to extract features, enabling the distinction of various types of defects. When compared with other deep learning models, CW-YOLO achieves higher detection accuracy while maintaining a detection rate of 103.58fps, meeting the requirements for real-time detection. Furthermore, the paper introduces the F1 score as a comprehensive evaluation metric, capturing both detection accuracy and recall rate. The results show that the F1 score is enhanced by approximately 10% after image fusion, demonstrating that the proposed method can significantly improve defect detection, particularly in cases involving difficult-to-distinguish defects like material shortages and scratches.
期刊介绍:
3D Printing and Additive Manufacturing is a peer-reviewed journal that provides a forum for world-class research in additive manufacturing and related technologies. The Journal explores emerging challenges and opportunities ranging from new developments of processes and materials, to new simulation and design tools, and informative applications and case studies. Novel applications in new areas, such as medicine, education, bio-printing, food printing, art and architecture, are also encouraged.
The Journal addresses the important questions surrounding this powerful and growing field, including issues in policy and law, intellectual property, data standards, safety and liability, environmental impact, social, economic, and humanitarian implications, and emerging business models at the industrial and consumer scales.