Multi-user Handwriting Identification System Based On Windows Phone

With the rapid development of information technology, The importance of identity authentication in the modern social activities has become increasingly obvious. Because of the problem of information security, traditional authentication methods cannot meet the needs of social development now. Biometric identity authentication can greatly improve the security and reliability.andwriting identification is a technology to distinguish the figure of a person. It has the characteristics of convenient acquisition, stable feature, high reliability, etc.Mobile phone now has become an necessary part of modern life, phone carries lots of personal information, mobile security problem is paid more and more attention, but the common biological characteristics are unsuitable for some kind of mobile phone. therefore, handwriting identification system based on the phone has a good prospect.The main content of this paper is to research the related algorithms about the handwriting identification, with the implement of the system on the mobile phone. The main contents are as follows:1. Firstly, introduce some detailed knowledge about the concept of handwriting identification and biometrics identification, then brief analysis of the realization of handwriting identification system is given here.and describe the handwriting acquisition process in detail.2. Preprocessing methods and feature extraction algorithm that used in this system are mainly introduced. The main effect of the pretreatment is to remove the redundant information in the image, to get better feature extraction. this paper uses two feature extraction methods, multi-layer segmentation feature and combination of direction feature and grid feature. When using the second method, the feature we get is in high dimension, here we use the Principal Component Analysis(PCA) to reduce redundant dimensions. In addition, we use Dynamic Time Warping(DTW) approach to dynamic programming of the feature vectors.3. Based on the above algorithm, using the VS2010we developed a multi-user handwriting identification software on the Windows Phone platform. It can achieve multi-user’s handwriting identification.