HandiText: handwriting recognition based on dynamic characteristics with incremental LSTM

The Internet of Things (IoT) is is a new manifestation of data science. In order to ensure the credibility of data about IoT devices, authentication has gradually become an important research topic in IoT ecosystem. However, traditional graphical passwords and text passwords can cause serious memory burdens. Therefore, a convenient method for determining user identity is needed. In this paper, we propose a handwriting recognition authentication scheme named HandiText based on behavior and biometrics features. When people write a word by hand, HandiText capture their static biological features and dynamic behavior features during the writing process (writing speed, pressure, etc.). The features are related to habits, which make it difficult for attackers to imitate. We also carry out algorithms comparisons and experiments evaluation to prove the reliability of our scheme. The experiment results show that the Long Short-Term Memory (LSTM) has best classification accuracy, reaching 99%, while keeping relatively low false positive rate and false negative rate. We also test other dataset, the average accuracy of HandiText reach 98%, with strong generalization ability. In addition, the 324 users we investigated indicated that they are willing to use this scheme on IoT devices.

[1]  Fei Yin,et al.  CASIA Online and Offline Chinese Handwriting Databases , 2011, 2011 International Conference on Document Analysis and Recognition.

[2]  Alain Rakotomamonjy,et al.  Variable Selection Using SVM-based Criteria , 2003, J. Mach. Learn. Res..

[3]  Albrecht Schmidt,et al.  SmudgeSafe: geometric image transformations for smudge-resistant user authentication , 2014, UbiComp.

[4]  Yongwha Chung,et al.  Performance Analysis of Smart Card-Based Fingerprint Recognition For Secure User Authentication , 2001, I3E.

[5]  Luiz Eduardo Soares de Oliveira,et al.  Ensemble of classifiers for off-line signature verification , 2008, 2008 IEEE International Conference on Systems, Man and Cybernetics.

[6]  Rick Wash,et al.  Understanding Password Choices: How Frequently Entered Passwords Are Re-used across Websites , 2016, SOUPS.

[7]  Lucila Ohno-Machado,et al.  Logistic regression and artificial neural network classification models: a methodology review , 2002, J. Biomed. Informatics.

[8]  Jürgen Schmidhuber,et al.  Applying LSTM to Time Series Predictable through Time-Window Approaches , 2000, ICANN.

[9]  L. Deng,et al.  The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web] , 2012, IEEE Signal Processing Magazine.

[10]  Hyeonjoon Moon,et al.  Biometrics Person Authentication Using Projection-Based Face Recognition System in Verification Scenario , 2004, ICBA.

[11]  Dawn Song,et al.  Smart Locks: Lessons for Securing Commodity Internet of Things Devices , 2016, AsiaCCS.

[12]  Nikita Borisov,et al.  The Tangled Web of Password Reuse , 2014, NDSS.

[13]  Ping Wang,et al.  Offline Dictionary Attack on Password Authentication Schemes Using Smart Cards , 2013, ISC.

[14]  Li Zhen.  Comparative studies on handwriting features of Chinese and English scripts , 2007 .

[15]  Taekyoung Kwon,et al.  Covert Attentional Shoulder Surfing: Human Adversaries Are More Powerful Than Expected , 2014, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[16]  Kuldip Singh,et al.  Cryptanalysis of password authentication schemes: Current status and key issues , 2009, 2009 Proceeding of International Conference on Methods and Models in Computer Science (ICM2CS).

[17]  Zhaohui Wu,et al.  Liveness Detection for Face Recognition , 2008 .

[18]  Zhi-Li Zhang,et al.  Multi-touch Authentication Using Hand Geometry and Behavioral Information , 2017, 2017 IEEE Symposium on Security and Privacy (SP).

[19]  Reza Safabakhsh,et al.  Offline text-independent writer identification using codebook and efficient code extraction methods , 2013, Image Vis. Comput..

[20]  G. O. Williams.  Iris recognition technology , 1997 .

[21]  Patrick J. Flynn,et al.  Comments on the CASIA version 1.0 Iris Data Set , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  Bogdan Hoanca,et al.  Real-time continuous iris recognition for authentication using an eye tracker , 2012, CCS '12.

[23]  Rolf Ingold,et al.  Combined Handwriting and Speech Modalities for User Authentication , 2009, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[24]  Horst Bunke,et al.  The IAM-database: an English sentence database for offline handwriting recognition , 2002, International Journal on Document Analysis and Recognition.

[25]  Johnny Long,et al.  No Tech Hacking: A Guide to Social Engineering, Dumpster Diving, and Shoulder Surfing , 2008 .

[26]  Ranga Rodrigo,et al.  TextCaps: Handwritten Character Recognition With Very Small Datasets , 2019, 2019 IEEE Winter Conference on Applications of Computer Vision (WACV).

[27]  Yu Chen,et al.  Biometric Authentication Under Threat : Liveness Detection Hacking , 2019 .

[28]  Paul C. van Oorschot,et al.  Revisiting Defenses against Large-Scale Online Password Guessing Attacks , 2012, IEEE Transactions on Dependable and Secure Computing.

[29]  Ted Taekyoung Kwon,et al.  TinyLock: Affordable defense against smudge attacks on smartphone pattern lock systems , 2014, Comput. Secur..