硅酸盐学报

基于机器视觉古陶瓷无损分类识别

作者：翁政魁1 管业鹏1 2 罗宏杰3

单位：1. 上海大学通信与信息工程学院上海 200444 2. 新型显示技术及应用集成教育部重点实验室上海 200072 3. 上海大学材料科学与工程学院上海 200444

分类号：TP391.7

出版年,卷(期):页码：2017,45(12):1833-1842

DOI：10.14062/j.issn.0454-5648

摘要：

为客观、有效地对古陶瓷进行无损分类，提出了一种基于机器视觉古陶瓷无损分类识别方法。通过遍历古陶瓷器型边缘轮廓，获取古陶瓷器型结构细节特征，并在HSI 空间下提取古陶瓷釉色多通道颜色直方图特征。同时，提取反映古陶瓷纹理多样性的LBP 纹饰特征。基于上述特征，采用机器学习方法实现古陶瓷器型结构、釉色及其纹饰图案的无损分类识别。结果表明：通过机器视觉可以有效地对古陶瓷进行分类识别；在以16 为曲率步长、9 为LBP 算子分块数时，分别提取古陶瓷结构，纹饰特征有较好的识别精度，其中，基于结构与釉色融合特征相比单一特征具有更好的识别效果；当古陶瓷发生结构或纹饰上的小部分缺损时，该方法可以保持一定的鲁棒性，当信息丢失或缺损为5%时，平均识别率依旧可达85%以上，可期望实现古陶瓷科技鉴定中的良好应用。

A classification method was proposed for non-destructive authentication of ancient ceramic based on machine vision to make the ancient ceramic identification more objective and accurate. The edge contour of ancient ceramic was traversed at first. The detail feature for the shape structure of ancient ceramic was determined by the corresponding curvature. A multi-channel color

histogram in HSI space was proposed to describe the glaze colors. Meanwhile, the LBP texture feature was extracted to reflect the colorful graphic pattern of ancient ceramic. Machine learning was used to perform classification and identification for non-destructive authentication of ancient ceramic in the features mentioned above. The experimental result shows that the ancient ceramic can be effectively identified by machine vision and set 16 as the curvature step, 9 as LBP operator block numbers during the extraction of structural and glaze feature outperforms than other parameters. Also, the fusion of structural and glaze features obtain a better recognition accuracy, compared to the single feature. The method can maintain a relative robustness when a small part of the ceramic structure or glaze is damaged. The average recognition accuracy can still reach more than 85% when the information loss or defect is 5%, which could allow its application in ancient ceramic scientific identification.

基金项目：

国家自然科学基金(11176016；60872117)；高等学校博士学科点专项科研基金(20123108110014)资助。

作者简介：

第一作者：翁政魁(1991—)，男，博士研究生

参考文献：

[1] 李家治, 王昌燧. 中国古陶瓷科技鉴定的基础和现状[C]//国古陶瓷无损科学鉴定的方法、前景和意义研讨会, 北京, 2002: 17–24.LI Jiazhi, WANG Changsui. The basis and present situation of scientific and technical determination of ancient Chinese ceramics[C]// Methodology, Prospect & Significance on Non-destroyed Scientific Determination for Chinese Ancient Ceramics-proceedings of Ccast (in Chinese), Beijing, 2002: 17–24.

[2] 王昌燧, 李家治. 关于建立中国古陶瓷科技鉴定中心的初步思考[C]//中国古陶瓷无损科学鉴定的方法、前景和意义研讨会, 北京,2002: 25–27.WANG Changsui, LI Jiazhi. Preliminary thoughts on establishment of China’s ancient ceramics science and technology determination center[C]//Methodology, Prospect & Significance on Non-destroyed Scientific Determination for Chinese Ancient Ceramics-proceedings of

Ccast (in Chinese), Beijing, 2002: 25–27.

[3] 王维达, 夏君定, 周智新. 热释光前剂量饱和指数法测定中国古瓷器年代[J]. 中国科学:技术科学, 2006, 36(5): 525–540.WANG Weida, XIA Junding, ZHOU Zhixin. Sci China Tech Sci (in Chinese), 2006, 36(5): 525–540.

[4] 杨云, 郭立文. 中国古陶瓷器型结构数据库的建立[J]. 陶瓷学报,2005, 26(1): 53–56.YANG Yun, GUO Liwen. J Ceram (in Chinese), 2005, 26(1): 53–56.

[5] 李卫民. 陶瓷鉴定理论的方法论研究[J]. 中国文物科学研究, 2011,5(1): 60–62. LI Weimin. China Cult Herit Sci Res(in Chinese), 2011, 5(1): 60–62.

[6] 张茂林, 吴军明, 李其江, 等. 古陶瓷科技鉴定法[J]. 文物鉴定与鉴赏, 2010, 2(4): 45–49. ZHANG Maolin, WU Junming, LI Qijiang, et al. Identif Appr Cult Relics (in Chinese), 2010, 2(4): 45–49.

[7] ELSAYED K, IMAM H, MADKOUR F, et al. Laser-induced breakdown spectroscopy technique in identification of ancient ceramics bodies and glazes[C]//Proceedings of 2nd International Conference on Modern Trends in Physics Research, Cairo, 2015: 187–199.

[8] ANDERSON L, CUNNINGHAM C, LINDSTROM R, et al. Identification of lead and other elements in ceramic glazes and housewares by 109Cd-induced X-ray fluorescence emission spectrometry[J]. J AOAC Int, 1995, 78(2): 407–412.

[9] 罗宏杰, 杨云, 王芬, 等. 不同历史时期耀州窑碗器型结构特征之研究[J]. 中国陶瓷工业, 2003, 10(6): 1–4.LUO Hongjie, YANG Yun, WANG Fen, et al. China Ceram Ind (in Chinese), 2003, 10(6): 1–4.

[10] WU Jun, XIONG Lu, TANG Min, et al. Identification of ancient ceramics by digital shape characterization[J]. Sci China Tech Sci, 2012, 55(9): 2441–2446.

[11] 熊露, 唐敏, 李其江, 等. 数字化碗类器型结构及其应用[J]. 陶瓷学报, 2015, 36(4): 414–418.XIONG Lu, TANG Min, LI Qijiang, et al. J Ceram (in Chinese), 2015,36 (4): 414–418.

[12] 周少华, 付略, 梁宝鎏. 基于SOM 神经网络的古代青瓷聚类分析[J]. 中国科学: 技术科学, 2008, 38(7): 1089–1096.ZHOU Shaohua, FU Lue, LIANG Baoliu. Sci China Tech Sci(in Chinese), 2008, 38(7): 1089–1096.

[13] KAMPEL M, SABLATING R. Rule based system for archaeological pottery classification[J]. Pattern Recogn Lett, 2007, 28(6): 740–747.

[14] 付略, 周少华, 彭勃, 等. 基于最小二乘支持向量机算法的南宋官窑出土瓷片分类[J]. 硅酸盐学报, 2008, 36(8): 1183–1186.FU Lue, ZHOU Shaohua, PENG Bo, et al. J Chin Ceram Soc, 2008,

36(8): 1183–1186.

[15] KADHEM A, ASADI T, HADI I. Determine blocks of image forhiding operation based on quad Chain code and DCT[J]. EAR, 2015,11(10): 12678–12689.

[16] BENDU H, DEEPAK B L, MURUGAN S. Application of GRNN for the prediction of performance and exhaust emissions in HCCI engine using ethanol[J]. Energ Convers Manage, 2016, 122(8): 165–173.

[17] 周永民. 宋瓷审美取向与庄子美学思想[J]. 中国陶瓷, 2008, 44(2):63–65.ZHOU Yongmin. Chin Ceram(in Chinese), 2008, 44(2): 63–65.

[18] JIANG H Q, ZHAO Y L, GAO J M, et al. Adaptive pseudo-color enhancement method of weld radiographic images based on HSI color space and self-transformation of pixels[J]. Rev Sci Instr, 2017, 88(6): 065106.

[19] 王彩丽. 浅显的祝福图案丰富的文化内涵——中国古陶瓷纹饰内涵寓意分析[J]. 阜阳师范学院学报(社会科学版), 2003, 3(3): 70–71.WANG Caili. J Fuyang Teach Coll: Soc Sci Ed(in Chinese), 2003, 3(3):70–71.

[20] GUO Z, ZHANG L, ZHANG D. A completed modeling of local binary pattern operator for texture classification[J]. IEEE Trans Image Process,2010, 19(6): 1657–1573.

[21] 故宫博物院: [EB/OL]. [2017–04–20]. http://www.dpm.org.cn/ pavilion/23459.html.

[22] 中国国家博物馆: [EB/OL]. [2017–04–20].http://www.chnmuseum.cn.

服务与反馈：

【文章下载】【加入收藏】