CAJ | 학술논문

在iOS平台上开发了一款人脸识别系统，借助OpenCV函数库实现了基于Haar-like特征和AdaBoost算法的人脸检测。提出了主成分分析和线性判别分析相结合的人脸识别算法，既避免了主成分分析方法对图像信息不分主次、忽视类别信息的缺陷，又降低了线性判别分析算法高运算量导致的大误差、小样本的局限性。实验结果表明该系统的识别效果良好。
재iOS평태상개발료일관인검식별계통，차조OpenCV함수고실현료기우Haar-like특정화AdaBoost산법적인검검측。제출료주성분분석화선성판별분석상결합적인검식별산법，기피면료주성분분석방법대도상신식불분주차、홀시유별신식적결함，우강저료선성판별분석산법고운산량도치적대오차、소양본적국한성。실험결과표명해계통적식별효과량호。
A face recognition system based on the iOS platform is developed. The face detection is realized based on Haar-like features and AdaBoost algorithm by using OpenCV Library. The face detection algorithm combining the Principal Component Analysis(PCA)and Linear Discriminate Analysis(LDA)methods is proposed. The proposed algorithm can not only overcome drawbacks of the conventional PCA, which deals with the image regardless of primary or secondary information and without making use of the class information, but also decrease the limitation of large errors and small samples due to the large amount of computations in the LDA. Experiment results show that the designed system has good face recognition efficiency.