Research On Iris Recognition Algorithms Based On Elastodynamics

The increasingly vigorous Internet environment drives and promotes the development of convenient and secure biometric technology.In the existing biometric recognition,iris recognition technology is the most potential biometric technology,both in recognition algorithm and application research.The main research contents of this paper are as follows:In the aspect of iris location,this paper locates the iris according to the different characteristics of the inner and outer edges.For the pupil edge,according to the gray characteristics of the iris image,the threshold method is used to binarize the image,then the mathematical morphology operation is used to remove the noise,and then the separated pupil is obtained,and then the edge is extracted by canny operator.Finally,hough transform is used for edge location.On this basis,the eyelid position is located to eliminate the influence of eyelid on the location of iris outer boundary,and then hough transform is used to locate the outer edge of iris to achieve fast and accurate location.In order to solve the problems of large computation and long positioning time in the traditional positioning method,the computational complexity of the algorithm in this paper is greatly reduced,and the speed and accuracy of boundary positioning are improved.In the aspect of feature normalization,an iris normalization method based on elastic dynamics is proposed in this paper.The iris adjusts the pupil size through the radial scaling of the elastic tissue structure,thus controlling the amount of light entering the eyeball.In iris recognition algorithm,it is necessary to compensate the radial deformation of iris.Based on the elastic tissue structure of iris,an iris normalization method with unequal elastic dynamic motion is proposed in this paper.By forming the expansion of elastic tissue to achieve pupil size scaling,experiments show that this method can better represent iris deformation.In the aspect of feature extraction and coding,combined with iris texture features,log-gabor filter is used for feature extraction in order to obtain local image texture information,and the iris feature template is formed after quantitative coding.Finally,the similarity of iris feature template is calculated by Hamming distance,and the recognition is completed.This method effectively avoids blindness in iris matching process and improves the efficiency and accuracy of recognition.Finally,the experimental results are analyzed,and the iris inner and outer boundary location algorithm proposed in this paper makes the boundary location more accurate and prepares for the subsequent recognition process.At the same time,the normalization model of elastic dynamics inequality proposed in this paper is more in line with the biological tissue structure of iris,and the experimental results show that the recognition accuracy of this algorithm is much higher than that of the traditional linear normalization algorithm.Finally,the log-gabor filtering feature extraction method not only ensures the robustness of the algorithm,but also greatly improves the matching success rate,reduces the error matching rate,and improves the operation efficiency,which provides a guarantee for the implementation of fast encryption and decryption.