基于改进遗传算法的虹膜识别方法研究

摘要

为了提高虹膜识别算法的有效性，提出了基于改进遗传算法进行特征选择的虹膜识别方法。针对虹膜图像纹理分布的特点，采用了多尺度环对称Gabor滤波器进行特征提取。同时，在特征选择上对经典遗传算法进行了改进，调整了遗传算法的交叉机制，提高了算法的搜索效率。为了检验算法的有效性，在中国科学院CASIA虹膜数据库上进行验证，实验结果表明该方法具有较高的识别精准度。%In order to improve the validity of iris recognition algorithm , the iris recognition method based on improved genetic algorithm for feature selection was proposed .Aiming at texture distribution characteristics of iris image , the multi-scale ring symmetric Gabor filter was used for feature extraction .At the same time , the classi-cal genetic algorithm was improved in the feature selection , the crossover mechanism of genetic algorithms was adjusted that improve the search efficiency of algorithm .In order to test the validity of the algorithm , the algo-rithm was verified in the CASIA iris database of Chinese Academy of Sciences .Experimental results show that our method has high recognition accuracy .