The Research Of Consumer Reward System Based On Palmprint Recognition Technology

As the rapid development of information and network technology, people’s demand for information security is increasingly improving. Biometric has been widely used because it is difficult to be copied with its inherent uniqueness. Specifically, biometric identification technology combines computer and biometric and uses the body’s unique physical characteristics or behavioral traits to identify the individual’s identity. It is easy to be carried and used, so it’s safe and reliable. In addition, that integrating biometric identification technology in monitoring, criminal investigation and other fields has bright prospect for social development.As a member of biometric, palmprint has been attracting the attention of many researchers in recent years with its rich information, stable and obvious features. Palmprint image’s capture equipment is simple and cheap and method is easy to be accepted by people; Recognition system is widely applied. Due to these advantages, palmprint recognition technology increasingly becomes that who has the great potential to develop in biometric recognition. The key issue of a successful palmprint recognition approach is how to extract the effective feature from a palmprint image and use the recognition technology to establish actual system. In this thesis, it establishes a palmprint capture system and acquires palmprint images and finally establishes a small-scale palmprint database; Besides, it combines the Gabor wavelets with principal component analysis to extract palmprint features, which has good recognition accuracy; Lastly it designs and implements a consumer reward system based on palmprint recognition. The main research contents in this thesis are listed as follows:(1) In the thesis, a palmprint capture equipment is established. CCD cameras are used to acquire palmprint images, which will be processed by DM642EVM development board and computer. Lots of palmprint images acquired by capture equipment will be stored in the palmprint database with corresponding user’s information.(2) The method of combining Gabor wavelets with principal component analysis is used to extract features of palmprint images that has been preprocessed. Firstly, The arithmetic makes use of Gabor wavelets transform to obtain Gabor eigenvector, but it has high dimension and is difficult to be calculated; Then the2-dimensional principal component analysis is used to reduce dimensions, but the arithmetic brings in many feature coefficients during the process, which makes it difficult to store and match; Finally, the1-dimensional principal component analysis is used to extract features from the eigenvector whose dimension has been reduced. The arithmetic not only makes use of good positioning properties of Gabor wavelets in frequency domain, but also effectively solves the problem of eigenvector’s high dimension. The experiment in the Polyu palmprint database and self-captured palmprint database proved that the method has good recognition accuracy and can effectively extracts features from palmprint images.(3) A consumer reward system is designed and implemented based on previous studies in palmprint recognition technology. The system can indentifies or enrolls palmprint images captured from users. Then according to the results of recognition, it decides whether the scores of user’s consumption will be cumulated. Besides, lots of interfaces as main windows, enrolling and indentifying for users are designed and implemented. Those interfaces are sample and easy to be operated for users.