2025 01 v.55 129-138
改进UNet++的瓷器文物显微气泡分割
基金项目(Foundation):
国家自然科学基金(62271393);
陕西省教育厅一般项目(19JK0842);
虚拟现实技术与系统全国重点实验室(北京航空航天大学)开放课题基金(VRLAB2024C02)
邮箱(Email):
ghgeng@nwu.edu.cn;
DOI:
10.16152/j.cnki.xdxbzr.2025-01-011
中文作者单位:
西北大学文化遗产数字化国家地方联合工程研究中心;西北大学可视化技术研究所;
摘要(Abstract):
对瓷器文物显微气泡的分割，可以更加清晰地观察瓷器表面微观气泡的形态、数量以及分布规律，进而辅助文物专家进行瓷器碎片分类和文物鉴定等工作。但瓷器显微图像中气泡复杂多变，大小及分布不均匀，现有图像分割方法难以适应瓷器显微气泡特征。因此，该文提出一种基于卷积激活单元的网络AGUNet++,该网络重新设计密集跳跃连接，节点间采用Z字形连接方式，充分提取图像语义特征，防止信息丢失；同时，在卷积单元的密集跳跃连接处，结合注意力门控模块Attention Gate提出卷积激活单元CAU,增强与瓷器文物显微气泡分割任务相关的气泡区域学习，抑制不相关的区域；在训练过程中对每一层子网络的输出采用深度监督和交叉熵损失，有效增强瓷器文物显微气泡特征提取能力，细化分割结果。该方法在SD-saliency-900以及PRMI数据集上的实验结果表明，与经典图像分割网络相比，AGUNet++在MIoU、Precision、Recall和F1分数中均有一定的提升，表现出更好的分割效果。
关键词(KeyWords):
瓷器文物显微图像;显微图像分割;UNet++;注意力门
| 143 | 0 | 14 |
| 下载次数 | 被引频次 | 阅读次数 |
参考文献
[1] 耿国华，冯龙，李康，等.秦陵文物数字化及虚拟复原研究综述[J].西北大学学报(自然科学版),2021,51(5):710-721.GENG G H,FENG L,LI K,et al.A literature review on the digitization and virtual restoration of cultural relics in the Mausoleum of Emperor Qinshihuang[J].Journal of Northwest University(Natural Science Edition),2021,51(5):710-721.
[2] 李春龙，周明全，成欣，等.轴对称破碎文物的虚拟复原方法[J].计算机辅助设计与图形学学报，2006,18(5):620-624.LI C L,ZHOU M Q,GENG X,et al.Virtual restoration of axisymmetric relic fragments[J].Journal of Computer-Aided Design and Computer Graphics,2006,18(5):620-624.
[3] 王飘，耿国华，张雨禾.基于表面纹理特征定义的碎片拼接方法[J].激光与光电子学进展，2018,55(8):81012.WANG P,GENG G H,ZHANG Y H.Fragment splicing method based on surface texture characteristic[J].Laser & Optoelectronics Progress,2018,55(8):081012.
[4] 张雨禾，耿国华，魏潇然，等.基于形状骨架图匹配的文物碎片自动重组方法[J].自动化学报，2017,43(4):622-633.ZHANG Y H,GENG G H,WEI X R,et al.Reassembly of fractured fragments based on skeleton graphs matching[J].Journal of Automation,2017,43(4):622-633.
[5] LI R X,GENG G H,WANG X Z,et al.LBCapsNet:A lightweight balanced capsule framework for image classification of porcelain fragments[J].Heritage Science,2024,12(1):1-14.
[6] 律海明.浅析古陶瓷的几种鉴定方法[J].文物鉴定与鉴赏，2014(4):90-94.LYU H M.A brief analysis of several identification methods for ancient ceramics[J].Identification and Appreciation to Cultural Relics,2014(4):90-94.
[7] 汤辉.标准化技术在古陶瓷眼学鉴定中的应用[J].景德镇陶瓷，2020(5):8-11.TANG H.The application of standardization techniques in visual identification of ancient porcelain[J].Jingdezhen Ceramics,2020(5):8-11.
[8] 金信苗.微观痕迹在古陶瓷鉴定中的有效性探索[J].科技风，2022(32):145-147.JIN X M.Exploration of the effectiveness of microscopic traces in the authentication of ancient porcelain[J].Technology Wind,2022(32):145-147.
[9] 王克刚.基于数字图像特征的古瓷片分类研究[D].西安：西北大学，2009.
[10] 鱼跃华.破损文物碎片的深度学习分类方法研究[D].西安：西北大学，2022.
[11] 陆正杰，李纯辉，耿国华，等.基于多特征描述子自适应权重的文物碎片分类[J].激光与光电子学进展，2020,57(4):321-329.LU Z J,LI C H,GENG G H,et al.Classification of cultural fragments based on adaptive weights of multi-feature descriptions[J].Laser & Optoelectronics Progress,2020,57(4):321-329.
[12] 耿国华，薛米妍，周蓬勃，等.基于对比学习与多尺度结合的陶瓷显微图像分类方法[J].西北大学学报(自然科学版),2021,51(5):734-741.GENG G H,XUE M Y,ZHOU P B,et al.Ceramic microscopic image classification based on the combination of contrastive learning and multi-scale methods[J].Journal of Northwest University(Natural Science Edition),2021,51(5):734-741.
[13] 刘景怡.基于显微特征的古代瓷器碎片分类方法的研究与实现[D].西安：西北大学，2021.
[14] 张佳宏.论古瓷器的气泡[J].收藏界，2012(12):118-119.ZHANG J H.On the bubbles of ancient porcelain[J].Collection World,2012(12):118-119.
[15] 李庆武，王珺，盛惠兴，等.基于小波包分析的陶瓷显微图像处理[C]//第十二届全国信号处理学术年会论文集.苏州，中国：中国电子学会信号处理分会，2005:321-324.
[16] 朱顺龙.古陶瓷显微科技气泡测定研究初探：以越窑青瓷研究为例[J].中国文物科学研究，2008(1):38-41.ZHU S L.Preliminary exploration of microscopic bubble measurement in ancient ceramics:A case study of Yue kiln celadon porcelain[J].China Cultural Heritage Scientific Research,2008(1):38-41.
[17] 刘国高.基于快速曲波变换的陶瓷显微图像处理[J].中国陶瓷工业，2008,15(2):17-21.LIU G G.Ceramic microscopic image procession based on fast curvelet transform[J].China Ceramic Industry,2008,15(2):17-21.
[18] LONG J,SHELHAMER E,DARREL T.Fully convolutional networks for semantic segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,39(4):640-651.
[19] RONNEBERGER O,FISCHER P,BROX T.U-Net:Convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention.Munich,Germany:Springer International Publishing,2015:234-241.
[20] ZHOU Z W,SIDDIQUEE M M R,TAJBAKHSH N,et al.UNet++:A nested U-Net architecture for medical image segmentation[J].Deep Learn Med Image Anal Multimodal Learn Clin Decis Support,2018,11045:3-11.
[21] SU L Z,XIE Q Y,ZHAO F J,et al.Change detection for multispectral images using modified semantic segmentation network[J].Journal of Applied Remote Sensing,2022,16(1):014518.
[22] ARDIYANTO I.Edge devices-oriented surface defect segmentation by ghostNet Fusion block and global auxiliary layer[J].Journal of Real-Time Image Processing,2024,21(1):13.
[23] LIU X D,WANG L L,HAN X G.Transformer with peak suppression and knowledge guidance for fine-grained image recognition[J].Neurocomputing,2022,492:137-149.
[24] WANG X L,GIRSHICK R,GUPTA A,et al.Non-local neural networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake,USA:IEEE,2018:7794-7803.
[25] CHEN L C,YANG Y,WANG J,et al.Attention to scale:Scale-aware semantic image segmentation[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas,USA:IEEE,2016:3640-3649.
[26] LUONG M T,PHAM H,MANNING C D.Effective approaches to attention-based neural machine translation[EB/OL](2015-08-17)[2024-02-14].https://arxiv.org/abs/1508.04025.
[27] BAHDANAU D,CHO K,BENGIO Y.Neural machine translation by jointly learning to align and translate[C]//3rd International Conference on Learning Representations.San Diego,USA:ICLR,2015:1-15.
[28] BRITZ D,GOLDIE A,LUONG M T,et al.Massive exploration of neural machine translation architectures[C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing.Copenhagen,Denmark:USAACL,2017:1442.
[29] MINAEE S,BOYKOV Y,PORIKLI F,et al.Image segmentation using deep learning:A survey[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,44(7):3523-3542.
[30] 苏虎，张家斌，张博豪，等.基于视觉感知的表面缺陷检测综述[J].计算机集成制造系统，2023,29(1):169-191.SU H,ZHANG J B,ZHANG B H,et al.A review of research on the inspection of surface defects based on visual perception[J].Computer Integrated Manufacturing Systems,2023,29(1):169-191.
[2] 李春龙，周明全，成欣，等.轴对称破碎文物的虚拟复原方法[J].计算机辅助设计与图形学学报，2006,18(5):620-624.LI C L,ZHOU M Q,GENG X,et al.Virtual restoration of axisymmetric relic fragments[J].Journal of Computer-Aided Design and Computer Graphics,2006,18(5):620-624.
[3] 王飘，耿国华，张雨禾.基于表面纹理特征定义的碎片拼接方法[J].激光与光电子学进展，2018,55(8):81012.WANG P,GENG G H,ZHANG Y H.Fragment splicing method based on surface texture characteristic[J].Laser & Optoelectronics Progress,2018,55(8):081012.
[4] 张雨禾，耿国华，魏潇然，等.基于形状骨架图匹配的文物碎片自动重组方法[J].自动化学报，2017,43(4):622-633.ZHANG Y H,GENG G H,WEI X R,et al.Reassembly of fractured fragments based on skeleton graphs matching[J].Journal of Automation,2017,43(4):622-633.
[5] LI R X,GENG G H,WANG X Z,et al.LBCapsNet:A lightweight balanced capsule framework for image classification of porcelain fragments[J].Heritage Science,2024,12(1):1-14.
[6] 律海明.浅析古陶瓷的几种鉴定方法[J].文物鉴定与鉴赏，2014(4):90-94.LYU H M.A brief analysis of several identification methods for ancient ceramics[J].Identification and Appreciation to Cultural Relics,2014(4):90-94.
[7] 汤辉.标准化技术在古陶瓷眼学鉴定中的应用[J].景德镇陶瓷，2020(5):8-11.TANG H.The application of standardization techniques in visual identification of ancient porcelain[J].Jingdezhen Ceramics,2020(5):8-11.
[8] 金信苗.微观痕迹在古陶瓷鉴定中的有效性探索[J].科技风，2022(32):145-147.JIN X M.Exploration of the effectiveness of microscopic traces in the authentication of ancient porcelain[J].Technology Wind,2022(32):145-147.
[9] 王克刚.基于数字图像特征的古瓷片分类研究[D].西安：西北大学，2009.
[10] 鱼跃华.破损文物碎片的深度学习分类方法研究[D].西安：西北大学，2022.
[11] 陆正杰，李纯辉，耿国华，等.基于多特征描述子自适应权重的文物碎片分类[J].激光与光电子学进展，2020,57(4):321-329.LU Z J,LI C H,GENG G H,et al.Classification of cultural fragments based on adaptive weights of multi-feature descriptions[J].Laser & Optoelectronics Progress,2020,57(4):321-329.
[12] 耿国华，薛米妍，周蓬勃，等.基于对比学习与多尺度结合的陶瓷显微图像分类方法[J].西北大学学报(自然科学版),2021,51(5):734-741.GENG G H,XUE M Y,ZHOU P B,et al.Ceramic microscopic image classification based on the combination of contrastive learning and multi-scale methods[J].Journal of Northwest University(Natural Science Edition),2021,51(5):734-741.
[13] 刘景怡.基于显微特征的古代瓷器碎片分类方法的研究与实现[D].西安：西北大学，2021.
[14] 张佳宏.论古瓷器的气泡[J].收藏界，2012(12):118-119.ZHANG J H.On the bubbles of ancient porcelain[J].Collection World,2012(12):118-119.
[15] 李庆武，王珺，盛惠兴，等.基于小波包分析的陶瓷显微图像处理[C]//第十二届全国信号处理学术年会论文集.苏州，中国：中国电子学会信号处理分会，2005:321-324.
[16] 朱顺龙.古陶瓷显微科技气泡测定研究初探：以越窑青瓷研究为例[J].中国文物科学研究，2008(1):38-41.ZHU S L.Preliminary exploration of microscopic bubble measurement in ancient ceramics:A case study of Yue kiln celadon porcelain[J].China Cultural Heritage Scientific Research,2008(1):38-41.
[17] 刘国高.基于快速曲波变换的陶瓷显微图像处理[J].中国陶瓷工业，2008,15(2):17-21.LIU G G.Ceramic microscopic image procession based on fast curvelet transform[J].China Ceramic Industry,2008,15(2):17-21.
[18] LONG J,SHELHAMER E,DARREL T.Fully convolutional networks for semantic segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,39(4):640-651.
[19] RONNEBERGER O,FISCHER P,BROX T.U-Net:Convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention.Munich,Germany:Springer International Publishing,2015:234-241.
[20] ZHOU Z W,SIDDIQUEE M M R,TAJBAKHSH N,et al.UNet++:A nested U-Net architecture for medical image segmentation[J].Deep Learn Med Image Anal Multimodal Learn Clin Decis Support,2018,11045:3-11.
[21] SU L Z,XIE Q Y,ZHAO F J,et al.Change detection for multispectral images using modified semantic segmentation network[J].Journal of Applied Remote Sensing,2022,16(1):014518.
[22] ARDIYANTO I.Edge devices-oriented surface defect segmentation by ghostNet Fusion block and global auxiliary layer[J].Journal of Real-Time Image Processing,2024,21(1):13.
[23] LIU X D,WANG L L,HAN X G.Transformer with peak suppression and knowledge guidance for fine-grained image recognition[J].Neurocomputing,2022,492:137-149.
[24] WANG X L,GIRSHICK R,GUPTA A,et al.Non-local neural networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake,USA:IEEE,2018:7794-7803.
[25] CHEN L C,YANG Y,WANG J,et al.Attention to scale:Scale-aware semantic image segmentation[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas,USA:IEEE,2016:3640-3649.
[26] LUONG M T,PHAM H,MANNING C D.Effective approaches to attention-based neural machine translation[EB/OL](2015-08-17)[2024-02-14].https://arxiv.org/abs/1508.04025.
[27] BAHDANAU D,CHO K,BENGIO Y.Neural machine translation by jointly learning to align and translate[C]//3rd International Conference on Learning Representations.San Diego,USA:ICLR,2015:1-15.
[28] BRITZ D,GOLDIE A,LUONG M T,et al.Massive exploration of neural machine translation architectures[C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing.Copenhagen,Denmark:USAACL,2017:1442.
[29] MINAEE S,BOYKOV Y,PORIKLI F,et al.Image segmentation using deep learning:A survey[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,44(7):3523-3542.
[30] 苏虎，张家斌，张博豪，等.基于视觉感知的表面缺陷检测综述[J].计算机集成制造系统，2023,29(1):169-191.SU H,ZHANG J B,ZHANG B H,et al.A review of research on the inspection of surface defects based on visual perception[J].Computer Integrated Manufacturing Systems,2023,29(1):169-191.
基本信息:
DOI：10.16152/j.cnki.xdxbzr.2025-01-011
中图分类号:K876.3;TP391.41
引用信息:
[1]刘阳洋,耿国华,刘鑫达等.改进UNet++的瓷器文物显微气泡分割[J].西北大学学报(自然科学版),2025,55(01):129-138.DOI:10.16152/j.cnki.xdxbzr.2025-01-011.
基金信息:
国家自然科学基金(62271393); 陕西省教育厅一般项目(19JK0842); 虚拟现实技术与系统全国重点实验室(北京航空航天大学)开放课题基金(VRLAB2024C02)