基于Gist和IPCA算法的多文种离线手写签名识别

计算机工程与科学 ›› 2022, Vol. 44 ›› Issue (11): 2048-2055.

基于Gist和IPCA算法的多文种离线手写签名识别

韩辉1,麦合甫热提2,3,吾尔尼沙·买买提2,朱亚俐1,库尔班·吾布力1,2

(1.新疆大学信息科学与工程学院，新疆乌鲁木齐 830046；2.新疆多语种信息技术重点实验室，新疆乌鲁木齐 830046；
3.新疆大学教务处，新疆乌鲁木齐 830046)

收稿日期:2021-04-08 修回日期:2021-08-08 接受日期:2022-11-25 出版日期:2022-11-25 发布日期:2022-11-25
基金资助:
国家自然科学基金(61862061,61563052,62061045)；新疆大学博士启动基金(BS180268)；新疆维吾尔自治区研究生科研创新项目(XJ2019G064,XJ2020G064)

Multilingual offline handwritten signature recognition based on Gist and IPCA

HAN Hui1,Mahpirat2,3,Hornisa Mamat2,ZHU Ya-li1,Kurban Ubul1,2

(1.School of Information Science and Engineering,Xinjiang University,Urumqi 830046;
2.Key Laboratory of Xinjiang Multilingual Information Technology,Urumqi 830046;
3.Office of Educational Administration,Xinjiang University,Urumqi 830046,China)

Received:2021-04-08 Revised:2021-08-08 Accepted:2022-11-25 Online:2022-11-25 Published:2022-11-25

摘要/Abstract

摘要： 由于离线手写签名图像有效的笔画部分普遍比较稀疏，存在大量的无效白色背景，目前常用的特征描述方法会使得得到的特征数据存在大量冗余，影响识别准确率。识别准确率的提高，需要依赖大量的训练数据和提取多个特征并进行融合，但这样又会因特征数据过多和维度过大而造成计算困难，影响识别效率。为此，提出了一种基于Gist和IPCA算法的多文种离线手写签名识别方法，利用Gist特征聚焦图像的整体布局和笔画部分，同时利用IPCA算法的批处理能力来提高识别效果和运行效率。使用中、英、维3种语言的实验数据集，并使用SVM分类器进行识别实验。结果显示，3个数据集上的识别准确率分别为97.97%,98.43%和97.19%，3种数据混合后的识别准确率为97.70%。经过对比分析可知，提出的方法与之前的相关方法相比明显较优。

关键词: 多文种, 手写签名识别, Gist特征, IPCA算法, SVM

Abstract: Because the effective strokes of offline handwritten signature images are generally sparse, and there are lots of invalid white backgrounds, using the commonly used feature description methods will cause a lot of re-dundancy in the obtained feature data, which will affect the recognition accuracy. In order to improve the recognition accuracy, we either need to rely on a large number of training data or extract multiple features for fusion, which will cause difficulty in the calculation and affect the efficiency of the experiment due to too much feature data and too large dimensions. Therefore, this paper proposes a multilingual off-line hand-written signature recognition method based on the Gist and IPCA algorithms, which uses gist features to focus on the overall layout and strokes of the image, and the batch processing ability of the IPCA algorithm to improve the recognition effect and operation efficiency. Three experimental datasets (Chinese, English, and Uyghur) and the SVM classifier are used in the recognition experiments. The results show that the recognition accuracy of the three data sets is 97.97%, 98.43%, and 97.19% respectively, and the recognition accuracy of the three mixed data sets is 97.7%. Comparative analysis shows that the proposal is obviously better than the previous related research.

Key words: multilingual, handwritten signature recognition, Gist feature, IPCA algorithm, SVM

韩辉, 麦合甫热提, 吾尔尼沙·买买提, 朱亚俐, 库尔班·吾布力, . 基于Gist和IPCA算法的多文种离线手写签名识别[J]. 计算机工程与科学, 2022, 44(11): 2048-2055.

HAN Hui, Mahpirat, Hornisa Mamat, ZHU Ya-li, Kurban Ubul, . Multilingual offline handwritten signature recognition based on Gist and IPCA[J]. Computer Engineering & Science, 2022, 44(11): 2048-2055.

参考文献［23］

［1］	Diaz M,Ferrer M A,Impedovo D,et al.A perspective analysis of handwritten signature technology［J］.ACM Computing Surveys,2019,51(6):1-39.
［2］	Alaei A,Pal S,Pal U,et al.An efficient signature verification method based on an interval symbolic representation and a fuzzy similarity measure［J］.IEEE Transactions on Information Forensics and Security,2017,12(10):2360-2372.
［3］	Ubul K, Adler A,Abliz G,et al.Off-line Uyghur signature recognition based on modified grid information features［C］∥Proc of the 11th International Conference on Information Science,Signal Processing and their Applications,2012:1056-1061.
［4］	Pal S,Alireza A,Pal U,et al.Multi-script off-line signature identification［C］∥Proc of the 12th International Conference on Hybrid Intelligent Systems,2012:236-240.
［5］	Serdouk Y,Nemmour H,Chibani Y.New gradient features for off-line handwritten signature verification［C］∥Proc of 2015 International Symposium on Innovations in Intelligent Systems and Applications,2015:1-4.
［6］	Liu Li-li.Off-line signature verification based on shape context［D］.Baoding:Hebei University,2016.(in Chinese)
［7］	Wei Jia-min,Feng Jun,Bu Qi-rong,et al.Off-line Chinese handwriting signature verification with hierarchical classification［J］.Computer Engineering and Applications, 2018,54(8):112-118.(in Chinese)
［8］	Ahat Yimin,Mutellip Mamut,Alimjan Aysa,et al.High- dimensional statistical features based Uyghur handwritten signature recognition［J］.Journal of Frontiers of Computer Science and Technology,2018,12(2):308-317.(in Chinese)
［9］	Mo L F,Zhu Y L,Mamat H,et al.Off-line handwritten signature recognition based on discrete curvelet transform［C］∥Proc of Chinese Conference on Biometric Recognition,2019:424-434.
［10］	Kushner T R,Rosenfeld A.A model of interprocessor communication for parallel image processing［J］.IEEE Transactions on Systems Man & Cybernetics,2012,SMC-13(4):600-618.
［11］	Friedman A. Framing pictures:The role of knowledge in automatized encoding and memory for Gist［J］.Journal of Experimental Psychology:General,1979,108(3):316-355.
［12］	Oliva A,Torralba A.Modeling the shape of the scene:A holistic representation of the spatial envelope［J］.International Journal of Computer Vision,2001,42(3):145-175.
［13］	Torralba A, Murphy K P,Freeman W T,et al.Context-based vision system for place and object recognition［C］∥Proc of IEEE International Conference on Computer Vision,2003:273-273.
［14］	Renninger L,Malik J.When is scene identification just texture recognition?［J］.Vision Research,2004,44(19):2301-2311.
［15］	Siagian C,Itti L.Rapid biologically-inspired scene classification using features shared with visual attention［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(2):300-312.
［16］	Chen Cong.Research on image scene recognition method considering scene complexity［D］.Nanjing:Nanjing Normal University,2017.(in Chinese)
［17］	Wang En-de,Liu Qiao-ying,Li Yong.Static human behavior classification based on LLC and GIST features［J］.Computer Engineering,2018,44(8):268-272.(in Chinese)
［18］	Stuttaford S A,Krasoulis A,Dupan S S G,et al.Automatic myoelectric control site detection using candid covariance-free incremental principal component analysis［C］∥Proc of the 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society,2020:3497-3500.
［19］	Chandra M A,Bedi S S.Survey on SVM and their application in image classification［J］.International Journal of Information Technology,2018(2):1-11.
［20］	Ferrer M A,Diaz-Cabrera M,Morales A.Static signature synthesis:A neuromotor inspired approach for biometrics［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(3):667-680.
［21］	Ferrer M A,Diaz-Cabrera M,Morales A.Synthetic off-line signature image generation［C］∥Proc of 2013 International Conference on Biometrics,2013:1-7.
［22］	Kurban Ubul,Rayima Ablikim,Nurbiya Yadikar,et al. Off-line Uyghur signature recognition technology based on density feature［J］.Computer Engineering and Design,2016,37 (8):2200-2205.(in Chinese)
［23］	Mo Long-fei,Mahpirat,Zhu Ya-li,et al.Off-line signature recognition based on non-downsampled contourlet transform［J］.Computer Engineering and Design,2020,41(12):180-186.(in Chinese)
	附中文参考文献：
［6］	刘利利.基于形状上下文的离线签名鉴别［D］.保定：河北大学,2016.
［7］	魏佳敏，冯筠，卜起荣，等.层次化分类的离线中文签名真伪鉴别方法［J］.计算机工程与应用，2018,54(8):112-118.
［8］	艾海提·伊敏,阿力木江·艾沙,库尔班·吾布力.高维统计特征融合的维吾尔文脱机手写签名识别［J］.计算机科学与探索,2018,12(2):308-317.
［16］	陈聪.顾及场景复杂度的图像自然场景识别方法研究［D］.南京：南京师范大学,2017.
［17］	王恩德,刘巧英,李勇.基于LLC与GIST特征的静态人体行为分类［J］.计算机工程,2018,44(8):268-272.
［22］	库尔班·吾布力,热依买·阿不力克木,努尔毕亚·亚地卡尔,等.基于密度特征的维吾尔文离线签名识别［J］.计算机工程与设计,2016,37(8):2200-2205.
［23］	莫龙飞,麦合甫热提,朱亚俐,等.基于非下采样轮廓波变换的离线签名识别［J］.计算机工程与设计,2020,41(12):180-186.