CAJ | 학술논문

针对典型相关分析（CCA）无法准确刻画虹膜图像的局部遮挡变化缺陷，提出一种改进典型相关分析相融合（ICCA）的虹膜识别方法。以全局和局部特征间的相关性特征作为有效的判别信息，通过划分子模，并以简单投票进行结果矫正，提高方法的稳定性，以CASIA数据集验证ICCA的有效性。结果表明，ICCA的识别率明显优于参比方法。
침대전형상관분석（CCA）무법준학각화홍막도상적국부차당변화결함，제출일충개진전형상관분석상융합（ICCA）적홍막식별방법。이전국화국부특정간적상관성특정작위유효적판별신식，통과화분자모，병이간단투표진행결과교정，제고방법적은정성，이CASIA수거집험증ICCA적유효성。결과표명，ICCA적식별솔명현우우삼비방법。
Canonical Correlation Analysis(CCA)can not better portray the local changes in the iris image, a novel iris recognition method is proposed based on improved CCA algorithm in this paper. Firstly, the correlation between global features and local features are integrated to form the recognition features, the redundant information between the features is eliminated and the global information and local information is integrated effectively at the same time, the performance of ICCA is tested by CASIA datasets. The result show that ICCA’s recognition accuracy is significantly better than the reference model.