中国刑事警察学院 刑事科学技术学院,辽宁 沈阳 110035
崔岚,硕士,教授,研究方向:文件检验,E-mail:cuilan0605@126.com
纸质出版日期:2024-04-15,
收稿日期:2023-11-14,
修回日期:2023-12-11,
扫 描 看 全 文
李硕,崔岚,付沛.高光谱喷墨打印墨水数据的非线性降维及分类建模方法研究[J].分析测试学报,2024,43(04):523-531.
LI Shuo,CUI Lan,FU Pei.Research on Nonlinear Dimensionality Reduction and Classification Modeling Methods of Hyperspectral Inkjet Printing Ink Data[J].Journal of Instrumental Analysis,2024,43(04):523-531.
李硕,崔岚,付沛.高光谱喷墨打印墨水数据的非线性降维及分类建模方法研究[J].分析测试学报,2024,43(04):523-531. DOI: 10.12452/j.fxcsxb.23111401.
LI Shuo,CUI Lan,FU Pei.Research on Nonlinear Dimensionality Reduction and Classification Modeling Methods of Hyperspectral Inkjet Printing Ink Data[J].Journal of Instrumental Analysis,2024,43(04):523-531. DOI: 10.12452/j.fxcsxb.23111401.
在法庭科学实践中,往往需要通过对文件中字迹墨水的成分分析来精确判定检材和样本文件的同一性。该文利用高光谱成像技术结合机器学习对喷墨打印墨水的种类进行区分,分别采集14套不同品牌、型号的4色(黑、青、品红和黄色)喷墨打印墨水打印的文件在400~1 000 nm范围的高光谱图像,共提取56种样品墨迹的光谱数据。使用均匀流形逼近与投影技术(UMAP)和T分布随机近邻嵌入技术(t-SNE)两种算法对高光谱喷墨打印墨水数据进行降维处理,然后建立极致梯度提升(XGBoost)、轻量级梯度提升机器学习(LightGBM)和支持向量机(SVM)3种分类模型,以1∶4的比例确定测试集和训练集,分别对原始数据和降维后的数据进行分类。实验结果显示,UMAP降维算法结合SVM模型对喷墨打印墨水分类的效果最优,黑色墨水样品的分类精度为90%左右,其余颜色墨水样品的分类精度均为100%。该研究为喷墨打印文件的检验鉴定提供了一种新的、无损、准确的鉴别方法。
In the practice of forensic science,it is often necessary to accurately determine the identity of the test material and the sample document by analyzing the composition of the ink in the document.Hyperspectral imaging technology combined with machine learning was used to distinguish the types of inkjet printing inks. Hyperspectral images of documents printed with 4 colors(black,blue,magenta and yellow) of 14 sets of different brands and models were collected in the range of 400-1 000 nm,and spectral data of 56 samples were extracted.Use the uniform manifold approximation and projection(UMAP) and T-distributed stochastic neighbor embedding(t-SNE) two algorithms for hyperspectral data dimension reduction processing inkjet printing ink,and then establish extreme gradient boosting(XGBoost),light gradient boosting machine(LightGBM)and support vector machine(SVM),determine the test set and training set in the ratio of 1∶4,and classify the original data and the data after dimensionality reduction respectively.The experimental results show that UMAP dimension reduction algorithm combined with SVM model has the best effect on the classification of inkjet printing inks. The classification accuracy of black ink samples is about 90%,and the classification accuracy of other color ink samples is 100%.This study provides a new,non-destructive and accurate identification method for inkjet printing documents.
高光谱成像技术喷墨打印墨水降维算法分类模型
hyperspectral imaging technologyinkjet printing inkdimensionality reduction algorithmclassification model
Kou J. Chin. J. Spectrosc. Lab.(寇瑾. 光谱实验室),2009,26(3):689-691.
Wang S C,Cui L,Song H,Xu S,Zhao P C,Chen X H. Phys. Test. Chem. Anal.:Chem. Anal.(王舒超,崔岚,宋辉,徐爽,赵鹏程,陈晓红. 理化检验-化学分册),2021,57(9):781-787.
Li G P,Zhang J Z,Liu S X,Liu S H,Cui L Y. Chin. J. Spectrosc. Lab.(李国平,张金庄,刘淑霞,刘世海,崔连义. 光谱实验室),2009,26(1):71-73.
Braz A,Lopez-Lopez M,Montalvo G,Ruiz C G. Aust. J. Forensic Sci.,2015,47(4):411-420.
Si F Q,Zhou H J,Jiang Y,Ye Q H,Yuan M Y,Zhao M J,Qian Y Y. Aerosp. Shanghai(Chinese & English)(司福祺,周海金,江宇,叶擎昊,袁牧野,赵敏杰,钱园园. 上海航天(中英文)),2023,40(3):93-98.
He L,Wan L,Gao H Y. Spectrosc. Spectral Anal.(贺露,万莉,高会议. 光谱学与光谱分析),2023,43(3):724-730.
Kendler S,Aharoni R,Cohen S,Raich R,Weiss S,Levy H,Mano Z,Fishbain B,Ron I. Forensic Sci. Int,2019,301:e55-e58.
Yang C M,Zhu Z B,Li Y C,Ma Y,Song H Y. Spectrosc. Spectral Anal.(杨春梅,朱赞彬,李昱成,马岩,宋海洋. 光谱学与光谱分析),2023,43(10):3266-3271.
Glomb P,Romaszewski M,Cholewa M,Domino K. Forensic Sci. Int.,2018,290:227-237.
El-Sharkawy Y H,Elbasuney S. Remote Sens. Appl.:Soc. Environ.,2019,13:31-38.
Wang S Y,Yang Y Z,He W W,Li R K. J. Instrum. Anal.(王书越,杨玉柱,何伟文,李润康. 分析测试学报),2021,40(10):1489-1496.
Wang S Y,He H Y,Lv R,He W W,Li C Y,Cai N B. J. Forensic Sci.,2022,67(2):550-561.
Liu M,Shen S,Wang N. Chin. J. Lumin. (刘猛,申思,王楠. 发光学报),2017,38(5):663-669.
Hu H Q,Wei Y P,Xu H X,Zhang L,Mao X B,Zhao Y P. Spectrosc. Spectral Anal.(胡会强,位云朋,徐华兴,张蕾,毛晓波,赵宇平. 光谱学与光谱分析),2023,43(6):1953-1960.
Devassy M B,George S. Forensic Sci. Int.,2020,311:110194.
Myasnikov E. 2020 International Multi-Conference on Industrial Engineering and Modern Technologies(FarEastCon). Vladivostok,Russia,IEEE,2020:1-5.
Maaten L,Hinton G. J. Mach. Learn. Res.,2008,9:2579-2605.
McInnes L,Healy J,Melville J. J. Open Source Software,2018,(29):861.
Chen T,Guestrin C. Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining,2016:785-794.
Ke G L,Meng Q,Finley T,Wang T,Chen W,Ma W,Ye Q,Liu T Y. Adv. Neural Inf. Process. Syst.,2017,30:3149-3157.
Cortes C,Vapnik V. Mach. Learn.,1995,20:273-297.
0
浏览量
45
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构