首页期刊信息编委及顾问期刊发行联系方式使用帮助常见问题ENGLISH
位置:首页 >> 正文
基于深度学习的混凝土孔结构图像法快速分析
作者:周双喜1  伟1  星1 魏永起2 喻乐华1 
单位:(1. 华东交通大学土木建筑学院 南昌 330013 2. 同济大学材料科学与工程学院 上海200092) 
关键词:孔结构 图像分析 混凝土 深度学习 
分类号:TU528
出版年,卷(期):页码:2019,47(5):0-0
DOI:
摘要:

 根据体视学原理,应用深度学习模型和计算机处理软件(Image-Pro Plus),针对混凝土内部气孔(>10 μm),介绍了基于深度学习的混凝土孔快速分析方法,并建立了应用于该分析方法混凝土孔结构数据集,计算出了单个截面需要的拍照数,混凝土试样为150,砂浆试样为80,满足了图像分析混凝土孔结构的实用性。结果表明:基于深度学习混凝土孔结构快速分析方法能够提高分析效率,具有良好操作性和典型代表性,适合深入研究混凝土孔结构与宏观性能的关联性;但在提高分析方法的精度方面还需要后续工作不断完善。

 A deep analysis method of concrete hole based on deep learning for concrete internal pores (>10 μm) was introduced according to the principle of stereology using a deep learning model and a software named Image-Pro Plus, and a data set of concrete pore structure applied to the analysis method was established. The photographing number of a single section was calculated. The number of photographs is 150 for concrete samples and 80 for mortar samples, which satisfies the practicality of image analysis of concrete pore structures. The results show that the feasibility and efficiency of concrete pore structure analysis via deep leaning can provide a facile and effective route for linking the microstructure to the macroscale performance of concrete. However, a further work is needed to continuously improve the accuracy of this analysis method.

 
基金项目:
作者简介:
参考文献:

 [1] 杨玉红, 李悦, 杜修力. 自密实混凝土早期自收缩及微观孔结构研究[J]. 建筑材料学报, 2010, 13(5): 601–606.

YANG Yuhong, LI Yue, DU Xiuli. J Build Mater(in Chinese), 2010, 13(5): 601–606. 
[2] 崔静洁, 何文, 廖世军, 等. 多孔材料的孔结构表征及其分析[J]. 材料导报, 2009, 23(13): 82–86.
Cui Jingjie, He Wen, Liao Shijun, et al. Mater Rev (in Chinese), 2009, 23(13): 82–86.
[3] PLEAU R, PIGEON M, LAURENCOT J L. Some findings on the usefulness of image analysis for determining the characteristics of the air-void system on hardened concrete[J]. Cem Concr Compos, 2001, 23(2–3): 237–246.
[4] BAUMGART C W, CAVE S P, LINDER K E. An Automated Image Processing System for Concrete Evaluation[C]. Office of Scientific & Technical Information Technical Reports, 1998.
[5] BAUMGART C, BLAIR R, LINDER K. Image Analysis of Hardened Concrete Performance Enhancement of the Automated Concrete Evaluation System (ACES)[J]. Concrete, 2006.
[6] JANA D. A round robin test on measurements of air void parameters in hardened concrete by various automated image analyses and ASTM C 457 methods[R]. Canada: 2007.
[7] PETERSON K. Air void analysis of hardened concrete via flatbed scanner[J]. J Transport Res Board, 2001// DOI: 10.3141/1775-06.
[8] 李启令. 定量显微学及其在水泥混凝土材料科学中的应用前景[J]. 建筑材料学报, 1990(4): 333–339. 
LI Qiling. J Build Mater (in Chinese), 1990(4): 333–339.
[9] 朱洪波, 闫美珠, 李晨,等. 图像分析宏观孔孔隙率对混凝土抗压强度的影响[J]. 建筑材料学报, 2015, 18(2): 275–280.
ZHU Hongbo, YANMeizhu, Li Chen, et al. J Build Mater (in Chinese), 2015, 18(2): 275–280.
[10] 张雄, 黄廷皓, 张永娟,等. Image-Pro Plus混凝土孔结构图像分析方法[J]. 建筑材料学报, 2015,18(1): 177–182.
ZHANG Xiong, HUANG Tingwei, ZHANG Yongjuan, et al. J Build Mater(in Chinese), 2015, 18(1): 177–182.
[11] 方永浩, 王锐, 庞二波, 等. 水泥–粉煤灰泡沫混凝土抗压强度与气孔结构的关系[J]. 硅酸盐学报, 2010, 38(4): 621–626.
FANG Yonghao, WANG Rui, Pang Erbo, et al. J Chin Ceram Soc, 2010, 38(4): 621–626.
[12] UNDERWOODE. Stereology and quantitative metallography[M]. New York: ASTM, 1975: 7–15
[13] 刘国权, 刘胜新, 黄启今,等. 金相学和材料显微组织定量分析技 术[J]. 中国体视学与图像分析, 2002, 7(4): 248–251. 
LIU Guoquan, LIUShengxin, HUANGQijin, et al.Chin J Stereology Image Anal(in Chinese), 2002, 7(4): 248–251.
[14] CHEN L C, ZHU Y, PAPANDREOU G, et al. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation [R/OL]. [2018-8-22]. https://arxiv.org/pdf/1802.02611.pdf.
[15] HOWARD A G.Some Improvements on DeepConvolutional Neural Network Based Image Classification[R/OL]. [2013-12-19]. https://arxiv.org/pdf/1312.5402/pdf.
[16] CHA Y J, CHOI W, Büyüköztürk O. Deep Learning‐Based Crack Damage Detection Using Convolutional Neural Networks[J]. Computer‐aided Civil & Infrastructure Engineering, 2017, 32(5): 361–378.
[17] RUSSELL B C, TORRALBA A, MURPHY K P, et al. LabelMe: A Database and Web-Based Tool for Image Annotation[J]. Inte J Computer Vision, 2008, 77(1/3): 157–173.
[18] ABADI M, BARHAM P, J CHEN, et al. Tensorflow: a system for large-scale machine learning[R/OL]. [2016-11-2]. https://www.usenix. org/system/files/conference/osdi16/osdi16-abadi.pdf.
[19] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs[J]. Computer Sci, 2014(4): 357–361.
[20] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs[J].IEEE Transac Pattern Anal Mach Intel, 2018, 40(4): 834–848.
[21] CHEN L C, PAPANDREOU G, SCHROFF F, et al. Rethinking Atrous Convolution for Semantic Image Segmentation[R/OL]. [2017-12-5]. https://arxiv.org/pdf/1706.05587.pdf.
[22] LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[C]// IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, 2015: 3431–3440.
[23] 郭冠英. 车速观测中最小样本数的确定[J]. 华东公路, 1987(5): 10–13.
GUO Guanying. East China Highway (in Chinese), 1987(5); 10–13.
[24] 李德成, 张桃林, Velde B., 等. 孔隙结构图像分析中不同试验因素对分析结果的影响[J]. 土壤学报, 2002, 39(1): 52–57.
LI Decheng, ZHANG Taolin, VELDE B., et al. J Soil Sci(in Chinese), 2002, 39(1): 52–57.
[25] 张建波. 混凝土孔隙分形特征表征氯离子渗透性能研究[D]. 中国建筑材料科学研究总院, 2010.
ZHANG Jianbo. Fractal characteristics of concrete pores to characterize chloride ion permeability[D]. China Building Materials Research Institute, 2010. (in Chinese)
[26] 金珊珊, 张金喜, 陈春珍, 等. 水泥砂浆孔结构分形特征的研究[J]. 建筑材料学报, 2011, 14(1): 92–97.
JIN Shanshan, ZHANG Jinxi, CHEN Chunzhen, et al. J Build Mater (in Chinese) 2011, 14(1): 92–97.
[27] 陈悦, 李东旭. 压汞法测定材料孔结构的误差分析[J]. 硅酸盐通报, 2006, 25(4): 198–201.
CHEN Yue, LIDongxu. Bull Chin Ceram Soc (in Chinese), 2006, 25 (4): 198–201.
[28] 沈业青, 邓敏, 莫立武. 孔结构测试技术及其在硬化水泥浆体孔结构表征中的应用[J]. 硅酸盐通报, 2009, 28(6): 1191–1196.
SHENYeqing, DENG Min, MO Liwu. Bull Chin Ceram Soc (in Chinese), 2009, 28(6): 1191–1196.
[29] 郑克仁, 蔡峰良, 周锡玲. 基于氮吸附的水泥-沥青复合硬化体孔结构分析[J]. 武汉理工大学学报, 2013, 35(2): 1–5.
ZHENG Keren, CAIFengliang, ZHOUXiling. J Wuhan Univ Technol (in Chinese), 2013, 35(2): 1–5.
[30] 胡曙光, 袁盼, 王发洲, 等. 背散射电子图像分析法在水泥基材料孔结构研究中的应用[J]. 建筑材料学报, 2017, 20(2): 316–320. 
HU Shuguang, YUAN Pan, WANG Fazhou, et al. J Build Mater(in Chinese), 2017, 20(2): 316–320.
[31] 主编单位南京水利科学研究院. 水工混凝土试验规程[M]// 中华人民共和国电力行业标准. 中国电力出版社, 2002: 267.
[32] ASTM C 457, Standard Test Method for Microscopical Determination of Parameters of Air -void System in Hardened Concrete[S].
[33] 张辉, 丁建彤, 高培伟. 混凝土气孔结构自动测试方法对比分析[J]. 低温建筑技术, 2009, 31(12): 7–9.
ZhANG Hui, DING Jianwei, GAO Peiwei. Low Temperature Arch Technol(in Chinese), 2009, 31(12): 7–9.
[34] 张辉, 高培伟, 石南南, 等. 混凝土气泡特征参数测试方法研究[J]. 红水河, 2009, 28(4): 50–53.
ZHANG Hui, GAO Peiwei, SHI Nannan, et al. Hongshui River (in Chinese), 2009, 28(4): 50–53.
 
服务与反馈:
文章下载】【加入收藏
中国硅酸盐学会《硅酸盐学报》编辑室
京ICP备10016537号-2
京公网安备 11010802024188号
地址:北京市海淀区三里河路11号    邮政编码:100831
电话:010-57811253  57811254    
E-mail:jccs@ceramsoc.com