摘要
为了提取电子笔迹的有效特征实现身份识别任务,提出将电子笔迹的各类特征进行色彩量化,形成笔痕点阵y-x形态特征图、坐标值及压力值的x/y/p-t时序变化特征图、运笔8方向特征图、运笔速度特征图及压力变化速度特征图;然后构建5层的卷积神经网络(CNN),并提出一种双向注意力机制捕获跨通道的特征交互信息,增加特征的聚合程度;最后,将电子笔迹的多特征图作为CNN网络的输入数据,并在不同的网络层进行融合识别。实验结果表明,所提出的方法显著加快网络收敛速度,并提高认证准确率和鲁棒性,12个人的识别正确率可达95%以上,50个人的识别正确率可达92%以上。电子笔迹多类特征的可视化图可用于辅助进行笔迹鉴定,融合识别方法应用于身份认证可规避个人隐私泄露的风险。
In order to extract effective features of electronic handwriting to achieve recognition tasks,it is proposed that various features of electronic handwriting are quantified by color to form the dot matrix y-x representation map of handwriting dat,x/y/p-t representation map of coordinate value and pressure value of p with time changing,8 direction feature map of handwriting,pen wielding velocity feature map and pen pressure velocity feature map.Then a 5-layer convolutional neural network(CNN)is constructed,and a 2-dimensional attention mechanism algorithm is proposed to capture cross-channel feature interaction information to increase the degree of feature aggregation.Finally,the multi-feature maps of electronic handwriting are used simultaneously as the input data of the CNN network,and the fusion recognition is carried out at different network layers.The experimental results show that the proposed method can significantly accelerate the network convergence speed and improve the recognition accuracy and robustness.The recognition accuracy rate for 12 people can reach more than 95%,and the accuracy rate for 50 people can reach more than 92%.The visual graphs of multi-features of electronic handwriting can be used to assist handwriting identification,and the fusion recognition method can avoid the risk of personal privacy disclosure.
作者
周娟
唐艳君
杨超
李云鹏
ZHOU Juan;TANG Yanjun;YANG Chao;LI Yunpeng(Intelligent Justice Engineering Research Center of Ministry of Education,Chongqing 400120,China;School of Criminal Investigation,Southwest University of Political Science and Law,Chongqing 400120,China)
出处
《中国人民公安大学学报（自然科学版）》
2024年第1期17-24,共8页
Journal of People’s Public Security University of China(Science and Technology)
基金
重庆市教委科学技术研究青年项目(KJQN201900307)
重庆市科学技术委员会科技项目(KJQN202000313)
关键词
电子笔迹
多特征
色彩量化
卷积神经网络
注意力机制
融合识别
electronic handwriting
multi-features
color quantitation
CNN
attention mechanism
fusion recognition
