灰度共生矩阵特征加权融合的维文签名鉴别  

Gray level co-occurrence matrix feature weighting fusion for Uyghur signature verification

作　　者：祖丽皮亚.艾尼 麦合甫热提 努尔毕亚.亚地卡尔 尤努斯.艾沙 库尔班.吾布力 Zulpiya Gheni;Mahpirat;Nuerbiya Yadikar;Yunus Aysa;Kurban Ubul(School of Information Science and Engineering,Xinjiang University,Urumqi 830046,China;Educational Administration Department,Xinjiang University,Urumqi 830046,China)

机构地区：[1]新疆大学信息科学与工程学院,新疆乌鲁木齐830046 [2]新疆大学教务处,新疆乌鲁木齐830046

出　　处：《计算机工程与设计》2018年第4期1195-1201,共7页Computer Engineering and Design

基　　金：国家自然科学基金项目(61563052;61163028);新疆维吾尔自治区高校科研计划创新团队基金项目(XJEDU2017T002)

摘　　要：提出灰度共生矩阵特征加权融合和BP神经网络的维吾尔文手写签名鉴别方法。提取灰度共生矩的能量、熵、惯性矩和局部平稳性等4种特征并对其进行加权融合的方法进行签名鉴别。实验结果表明,15名签名者的(共600个签名样本)平均签名鉴别率为91.78%。与同样规模的GPDS英文签名进行对比实验可知,本文提出的方法对维吾尔文手写签名具有较高的稳定性和鉴别能力。A Uyghur handwritten signature verification method based on Gray level co-occurrence matrix feature weighting fusion and BP neural networks was proposed.In the feature extraction stage,the energy,entropy,moment of inertia and local statio-nary parameters of the GLCM were extracted and the weighted fusion features were carriedout to verify the signature.The experimental results show that,it is obtained 91.78%average of verification rate for 15 Uyghur signers(total of 600 signature samples).The comparison of experimental results with the GPDS English signatures of same size proved that,the proposed method has higher stability and verification accuracy for Uyghur handwritten signature.

关 键 词：维吾尔文 离线签名鉴别 灰度共生矩阵 特征加权融合 BP神经网络 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]