Deep learning for enhanced porosity detection in AZ91 magnesium alloys using windowed perception and aggregated sensing

Abstract

In this study, we innovatively proposed a deep learning model architecture to address the industry challenges in the detection of porosity in magnesium alloys. Magnesium alloys, known for their lightweight and high-strength characteristics, are extensively utilized in aerospace, automotive, and biomedical fields. However, the absorption of hydrogen during the production process leads to the formation of pores, which not only reduce the material’s strength and durability but may also cause premature failure of the material. The formation of pores typically occurs during the solidification stage of magnesium alloys, where hydrogen dissolved in the molten metal is released upon cooling, forming tiny gas pores. The presence of these gas pores significantly affects the mechanical properties of the material, potentially leading to crack initiation and propagation under high stress. Therefore, accurate detection and quantification of pores are crucial for enhancing the quality control of magnesium alloys. Our developed model integrates window-shaped perception blocks with convolutional neural networks, enhanced by aggregated sensing layers (ASLs) on long-range connections. Extensive training on real samples demonstrated that our model outperforms mainstream algorithms such as U-Net and TransUNet across various evaluation metrics, particularly in fine target detection tasks under complex scenarios. Specifically, our model achieved a Dice coefficient of 74.77% and an Intersection over Union index of 71.00%, significantly surpassing other models. Moreover, the method also demonstrated superior accuracy in pore edge prediction, effectively mitigating issues of oversegmentation and undersegmentation, especially for small and irregular pores. An ablation study further confirmed the effectiveness of each component, with the ASL module showing particular strength in feature extraction and reducing upsampling loss. In summary, this research highlights the significant potential of deep learning technology in material defect detection and provides an efficient, automated solution for practical production, contributing to advancements in materials science and industrial quality control.

Keywords

AZ91 magnesium alloy / image segmentation / deep learning / porosity defects

References

[1] Zhang Y,Jiang B,Dong Z.Modeling the correlation between texture characteristics and tensile properties of AZ31 magnesium alloy based on the artificial neural networks.J Mater Res Technol2023;24:5286-97

[2] Bai J,Wen C.Applications of magnesium alloys for aerospace: a review.J Magnes Alloys2023;11:3609-19

[3] Tian P.Surface modification of biodegradable magnesium and its alloys for biomedical applications.Regen Biomater2015;2:135-51 PMCID:PMC4669019

[4] Yue X,Zhang M,Ma X.Additive manufacturing of high porosity magnesium scaffolds with lattice structure and random structure.Mater Sci Eng A2022;859:144167

[5] Abazari S,Bakhsheshi-Rad HR.Magnesium-based nanocomposites: a review from mechanical, creep and fatigue properties.J Magnes Alloys2023;11:2655-87

[6] Wang S,Xie D.Grain refinement and strength enhancement in Mg wrought alloys: a review.J Magnes Alloys2023;11:4128-45

[7] V K,Kumar SS.Magnesium role in additive manufacturing of biomedical implants - challenges and opportunities.Addit Manuf2022;55:102802

[8] Katunin A,Dragan K.Impact damage evaluation in composite structures based on fusion of results of ultrasonic testing and X-ray computed tomography.Sensors2020;20:1867 PMCID:PMC7180951

[9] Fu Y,Yuan L,Pratt A.Machine learning algorithms for defect detection in metal laser-based additive manufacturing: a review.J Manuf Process2022;75:693-710

[10] Minaee S,Porikli F,Kehtarnavaz N.Image segmentation using deep learning: a survey.IEEE Trans Pattern Anal Mach Intell2022;44:3523-42

[11] Li K,Qin Y.A review of the multi-dimensional application of machine learning to improve the integrated intelligence of laser powder bed fusion.J Mater Proc Technol2023;318:118032

[12] Gao G,Yu Y,Yang J.MSCFNet: a lightweight network with multi-scale context fusion for real-time semantic segmentation.IEEE Trans Intell Transport Syst2022;23:25489-99

[13] Hong D,Meng D,Chanussot J.Multimodal GANs: toward crossmodal hyperspectral–multispectral image segmentation.IEEE Trans Geosci Remote Sensing2021;59:5103-13

[14] Wang L,Zhang C.UNetFormer: a UNet-like transformer for efficient semantic segmentation of remote sensing urban scene imagery.ISPRS J Photogramm Remote Sens2022;190:196-214

[15] Lu W,Nguyen M.A lightweight CNN–transformer network with laplacian loss for low-altitude UAV imagery semantic segmentation.IEEE Trans Geosci Remote Sensing2024;62:1-20

[16] Wu J,Zhang H,Yang Q.Fault detection based on fully convolutional networks (FCN).JMSE2021;9:259

[17] Papadeas I,Amanatiadis A.Real-time semantic image segmentation with deep learning for autonomous driving: a survey.Appl Sci2021;11:8802

[18] Chu P,Lammers K,Liu X.Deep learning-based apple detection using a suppression mask R-CNN.Pattern Recognit Lett2021;147:206-11

[19] Yang J,Zhang X,Zheng X.TSE DeepLab: an efficient visual transformer for medical image segmentation.Biomed Signal Process Control2023;80:104376

[20] Lin K,Lv J.Face detection and segmentation based on improved mask R-CNN.Discrete Dyn Nat Soc2020;2020:1-11

[21] Qiong L,Jinnan T,Jianxiang S.Medical image segmentation based on frequency domain decomposition SVD linear attention.Sci Rep2025;15:2833 PMCID:PMC11754837

[22] Banjanovic-Mehmedovic L,Gurdic Ribic A,Karabegovi I.Advancements in robotic intelligence: the role of computer vision, DRL, transformers and LLMs. 2024.

[23] Kolides A,Rathor A.Artificial intelligence foundation and pre-trained models: fundamentals, applications, opportunities, and social impacts.Simul Model Pract Theory2023;126:102754

[24] Han D,Han Y,Huang G.Flatten transformer: vision transformer using focused linear attention. In: 2023 IEEE/CVF International Conference on Computer Vision (ICCV), Paris, France. Oct 01-06, 2023. IEEE, 2023; pp. 5961-71.

[25] Chen J,Li X. 3D TransUNet: advancing medical image segmentation through vision transformers. arXiv 2023, arXiv:2310.07781. Available online: https://doi.org/10.48550/arXiv.2310.07781. (accessed on 12 Mar 2025)

[26] Ozcan A,Donmez E.Enhanced-TransUNet for ultrasound segmentation of thyroid nodules.Biomed Signal Process Control2024;95:106472

[27] Jain J,Chiu M,Orlov N. OneFormer: one transformer to rule universal image segmentation. arXiv 2022, arXiv:2211.06220. Available online: https://doi.org/10.48550/arXiv.2211.06220. (accessed on 12 Mar 2025)

[28] Chen J,Li X.TransUNet: rethinking the U-Net architecture design for medical image segmentation through the lens of transformers.Med Image Anal2024;97:103280

[29] Chen J,Yu Q. TransUnet: transformers make strong encoders for medical image segmentation. arXiv 2021, arXiv:2102.04306. Available online: https://doi.org/10.48550/arXiv.2102.04306. (accessed on 12 Mar 2025)

[30] Anand V.PoreNet: CNN-based pore descriptor for high-resolution fingerprint recognition.IEEE Sensors J2020;20:9305-13

[31] Al-Zaidawi SMK.A pore classification system for the detection of additive manufacturing defects combining machine learning and numerical image analysis.Eng Proc2023;58:122

[32] Budd S,Kainz B.A survey on active learning and human-in-the-loop deep learning for medical image analysis.Med Image Anal2021;71:102062

[33] Zheng S,Leung T.Improving the robustness of deep neural networks via stability training. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, USA. Jun 27-30, 2016. IEEE, 2016; pp. 4480-8.

[34] Michaelis C,Geirhos R. Benchmarking robustness in object detection: autonomous driving when winter is coming. arXiv 2019, arXiv:1907.07484. Available online: https://doi.org/10.48550/arXiv.1907.07484. (accessed on 12 Mar 2025)

[35] Farrukh YA,Khan I.SeNet-I: an approach for detecting network intrusions through serialized network traffic images.Eng Appl Artif Intell2023;126:107169

[36] Huang Y,He H,Zhao B.Senet: spatial information enhancement for semantic segmentation neural networks.Vis Comput2024;40:3427-40

[37] Zhao Y,Huang L.SEF-UNet: advancing abdominal multi-organ segmentation with SEFormer and depthwise cascaded upsampling.PeerJ Comput Sci2024;10:e2238 PMCID:PMC11323136

[38] Cai Z,Wang G,Zhang X. Align-DETR: enhancing end-to-end object detection with aligned loss. arXiv 2023, arXiv:2304.07527. Available online: https://doi.org/10.48550/arXiv.2304.07527. (accessed on 12 Mar 2025)

[39] Nong X,Lin S,Ye X.Multimodal deep neural network-based sensor data anomaly diagnosis method for structural health monitoring.Buildings2023;13:1976

[40] Alammar Z,Zhang J,Lafta W.Deep transfer learning with enhanced feature fusion for detection of abnormalities in X-ray images.Cancers2023;15:4007 PMCID:PMC10417687

[41] Tang Q,Zhu F.A comparative review on multi-modal sensors fusion based on deep learning.Signal Process2023;213:109165

[42] Fang Y,Wu R.What makes for hierarchical vision transformer?.IEEE Trans Pattern Anal Mach Intell2023;45:12714-20