摘要
特征提取是图像目标处理分类和识别的关键。将像素信息表示的图像数据信息转换为特征向量,在减少数据量的同时保留图像中包含的原始信息。对主成分分析与二维主成分分析两种特征提取算法对比研究,提出一种改善的二维主成分分析图像特征提取算法。算法是基于二维度的图像矩阵,使用标准ORL人像数据库。实验结果表明该算法在效率、准确率上均优于主成分分析方法。
Image feature extraction has become the key to processing and sorting target image.The result of the feature extraction is to convert the image data information represented by the pixel information into a feature vector,so as to preserve the original information contained in the image while reducing the amount of data.The subject of this paper is a comparative study of two feature extraction algorithms: principal component analysis and two dimensional principal component analysis,put forward a kind of improved two dimensional principal component analysis of image feature extraction algorithms.The algorithm structure is based on a two dimensional image matrix directly.The experiment uses the standard ORL library as the face dataset.The experimental results show that the improved algorithm is superior to the principal component analysis in efficiency and accuracy.
作者
赵蔷
惠燕
张忠
刘咪
ZHAO Qiang;HUI Yan;ZHANG Zhong;LIU Mi(School of Computer Science,Xianyang Normal University,Xianyang 712000,China)
出处
《航空计算技术》
2019年第5期40-42,共3页
Aeronautical Computing Technique
基金
陕西省教育厅专项科研计划项目资助(15JK1803)
陕西省教育科学“十三五”规划2017年课题项目资助(SGH17H197)
陕西省2018年大学生创新创业训练计划项目资助(201828036)
