| 606 | 10 | 48 |
| 下载次数 | 被引频次 | 阅读次数 |
提出一种借助于人工智能的辅助鉴别方法鉴定具有蕉叶纹饰的古陶瓷年代及真伪,利用局部二值模式(LBP)和方向梯度直方图(HOG)技术从古陶瓷图像的蕉叶纹饰中提取古陶瓷的时代特征,通过蕉叶纹局部二值模式纹饰特征和方向梯度直方图纹饰特征的融合实现古陶瓷的智能断代,并采用纹饰相似度衡量古陶瓷的真、伪可能性。以博物馆现有馆藏古陶瓷图像资料为基础进行实验。结果表明:该方法不仅可以完成古瓷器准确断代,而且能够发现蕉叶纹饰的发展规律,从而强有力地辅助专家的断代工作。此外,基于纹饰特征的相似度分析能够对古陶瓷的真伪辨识提供较有力支撑。
Abstract：An assisted identification method with artificial intelligence was proposed to identify the dating and authenticity of ancient ceramics with banana leaf pattern. The local binary patterns(LBP) and the histogram of oriented gradient(HOG) were used to extract the characteristics of the time from the banana leaf pattern of ancient ceramics images. The intelligent chronological identification of ancient ceramics was realized through the fusion of the local binary patterns features and the histogram of oriented gradient features of banana leaf pattern, and the similarity of ornamentation was used to measure the authentic and imitative possibility of ancient ceramics. The experiment was based on the ancient ceramic image database of the museum collections. The experimental results show that the proposed method can complete the accurate identification of ancient ceramics and find out the development law of banana leaf pattern, thus assisting experts in the identification of the ancient ceramics. In addition, the similarity analysis based on ornamentation features can provide a support for the authenticity identification of ancient ceramics.
[1]吴晓萍,管业鹏,李伟东,等.可见-近红外光谱的古陶瓷断代分类识别[J].光谱学与光谱分析,2019,39(3):756-764.WU Xiaoping,GUAN Yepeng,LI Weidong,et al.Spectrosc Spect Anal (in Chinese),2019,39(3):756-764.
[2]陈显求.宋耀州兔毫天目瓷釉的分相与析晶[J].自然杂志,1995(6):329-331.CHEN Xianqiu.Nature(in Chinese),1995(6):329-331.
[3]周波,谢光.人工智能在模式识别中的关键技术探究[J].现代信息科技,2019,3(22):110-111.ZHOU Bo,XIE Guang.Mod Inf Technol(in Chinese),2019,3 (22):110-111.
[4]翁政魁,管业鹏,罗宏杰.基于机器视觉古陶瓷无损分类识别[J].硅酸盐学报,2017,45(12):1833-1842.WENG Zhengkui,GUAN Yepeng,LUO Hongjie.J Chin Ceram Soc(in Chinese),2017,45(12):1833-1842.
[5]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(6):1-4.LUO Hongjie,YANG Yun,WANG Fen,et al.Chin Ceram Ind(in Chinese),2003(6):1-4.
[6]吴隽,罗宏杰,李家治,等.中国古陶瓷的断源断代[J].硅酸盐学报,2007(S1):39-43.WU Jun,LUO Hongjie,LI Jiazhi,et al.J Chin Ceram Soc,2007,(S1):39-43.
[7]胡薰尹,管业鹏,李伟东,等.基于紫外-可见-近红外光谱特征映射矩阵的古陶瓷分类方法[J].硅酸盐学报,2018,46(9):1280-1286.HU Xunyin,GUAN Yepeng,LI Weidong,et al.J Chin Ceram Soc,2018,46(9):1280-1286.
[8]MU Tianhong,WANG Fen,WANG Xiufeng,et al.Research on ancient ceramic identification by artificial intelligence[J].Ceram Int,2019,45:18140-18146.
[9]吴隽,熊露,唐敏,等.古瓷类文物器型结构的数字化特征鉴定[J].中国科学:技术科学,2012,42(9):1097-1102.WU Jun,XIONG Lu,TANG Min,et al.Sci Chin Tech Sci(in Chinese),2012,42(9):1097-1102.
[10]HEIN I,DOMINGUEZB A R,ORNELAS M,et al.Automated classification of archaeological ceramic materials by means of texture measures[J].J Archaeol Sci:Rep,2018,21:921-928.
[11]刘乐君,胡旋尹.“瓷上蕉叶”--五彩瓷中蕉叶纹图式变迁[J].中国陶瓷,2016,52(9):101-104+107.LIU Lejun,HU Xuanyin.China Ceram(in Chinese),2016,52(9):101-104+107.
[12][瓷元素]瓷器纹饰之--蕉叶纹.[EB/OL].[2018-08-07].http://m.sohu.com/a/245817700_99940915?spm=smbd.content.0.0.15781011216598xpSMOY.
[13]古陶瓷鉴定研究--如何通过纹饰鉴定古陶瓷(三).[EB/OL].[2018-01-21].http://www.360doc.com/content/18/0121/15/40042320_723881557.shtml
[14]MAHALE Vivek H.,ALI Mouad M,YANNAWAR Pravin L,et al.Image Inconsistency Detection Using Local Binary Pattern (LBP)[J].Procedia Comput Sci,2017,115:501-508.
[15]PAN Zhibin,WU Xiuquan,LI Zhengyi.Central pixel selection strategy based on local gray-value distribution by using gradient information to enhance LBP for texture classification[J].Expert Syst Appl,2019,120:319-334.
[16]LEI Yiming,ZHAO Ximei,WANG Guodong,et al.A novel approach for cirrhosis recognition via improved LBP algorithm and dictionary learning[J].Biomed Signal Process Control,2017,38:281-292.
[17]LIU Peizhong,GUO Jingming,CHAMNONGTHAI Kosin,et al.Fusion of color histogram and LBP-based features for texture image retrieval and classification[J].Inf Scis,2017,390:95-111.
[18]WEI Yun,TIAN Qing,GUO Jianhua,et al.Multi-vehicle detection algorithm through combining Harr and HOG features[J].Math Comput Simulat,2019,155:130-145.
[19]SUN Haotong,LV Guodong,MO Jiaqing,et al.Application of KPCAcombined with SVM in Raman spectral discrimination[J].Optik,2019,184:214-219.
[20]ARABI Punal M,JOSHI Gayatri,DEEPA N Vamsha.Performance evaluation of GLCM and pixel intensity matrix for skin texture analysis[J].Persp Sci,2016,8:203-206.
[21]KIM Namchul,SO Hyunjoo.Directional statistical Gabor features for texture classification[J].Pattern Recogn Lett,2018,112:18-26.
[22]冯丹丹,袁筱清.科技创新断古陶瓷真伪[J].景德镇陶瓷,2010,20(4):11.FENG Dandan,YUAN Xiaoqing.Jingdezhen’s Ceram(in Chinese),2010,20(4):11.
基本信息:
DOI：10.14062/j.issn.0454-5648.20200032
中图分类号:K247;K248;TQ174.1
引用信息:
[1]张静,周强,王莹,等.元明清陶瓷蕉叶纹纹饰的特征[J].硅酸盐学报,2020,48(09):1373-1381.DOI:10.14062/j.issn.0454-5648.20200032.
基金信息:
陕西省科技计划项目(2019GY-090);; 咸阳市科技计划项目(2017K02-06)资助
2020-07-13
2020-07-13
2020-07-13