The Iris Recognition Algorithm, Based On Two-dimensional Gabor Transform

With the development of information technology, information security becomes an important and urgent problem gradually. Therefore, biological characteristic recognition, which can be used to protect the information, is attracting more and more attention. Iris recognition is a new kind of biological characteristic recognition. Compared with the other biological characteristic recognitions (fingerprint recognition, facial recognition, voice recognition, etc.), iris recognition has the following characteristics: high stability, high reliability and non-contact.Concentrating on increasing the recognition rate, this dissertation presents some new algorithms and new schemes. The main contents and innovative contributions of' this dissertation are as follows: This dissertation introduces the significance and aim of iris recognition research first, and reviews the basic theory, development and current research status of iris recognition technology, and points out the existent problems and shortcomings. After comparing the commonly texture analysis methods, this dissertation gives one new iris texture analysis method based Gray-Primitive Co-occurrence Matrix. The Matrix can be used to describe the gray statistical distribution and the local detail structure in iris image texture. The experimentation and results show the analysis method was easier and more effective to be realized in real time. Aiming at the shortcoming that conventional 2D-Gabor filters based global feature iris recognition method can not describe the local fleck feature well, this dissertation presents the improved 2D-Gabor filter based global feature iris recognition. The improved method makes use of the no-linear transform and weighted feature to increase the recognition rate. Otherwise, this dissertation extracts the orientation local energy feature to encode as the iris local feature. Simulations validate the new method. The fusion method combined the proposed improved global feature and orientation local energy feature is designed to advance the algorithm recognition rate and the robust. The experimentations and results show this fusion method has reduced the False Accept Rate (FAR) with the higher recognition rate. Good recognition results are achieved after recognition experiments have been done in the CASIA iris database (version 1.0).