摘要
叶分量分析是一种增量式子空间学习方法,在图像处理中能够快速对重要的特征进行提取及表示,比其他子空间算法具有更优异的性能。本文对LCA(叶分量分析方法)做了深入的剖析,通过对静态图片的特征提取以及在线学习的过程,提高了对静态图片的识别率,并用matlab进行多幅图像的识别。结果表明,采用LCA算法在线学习,在ORL标准人连库的图像识别率可达到85.5%,识别率较高。同时,此算法优点在于执行速度快且没有小样本的情况出现,在样本较少的情况下也能达到很高的识别率。
Leaf component analysis is an incremental subspace learning method,which can quickly extract and express important features in image processing,and has better performance than other subspace algorithms.In this paper,LCA(leaf component analysis method)has been deeply analyzed.Through the process of feature extraction of static pictures and online learning,the recognition rate of static pictures is improved,and matlab is used to recognize multiple images.The results show that using LCA algorithm online learning,the image recognition rate in the ORL standard human-linked library can reach 85.5%,which is a high recognition rate.At the same time,the advantage of this algorithm is that the execution speed is fast and there are no small samples,and a high recognition rate can be achieved with fewer samples.
作者
崔莹
CUI Ying(Department of Information Engineering,Tongling Vocational and Technical College,Tongling Anhui 244000,China)
出处
《佳木斯大学学报(自然科学版)》
CAS
2022年第4期55-58,共4页
Journal of Jiamusi University:Natural Science Edition
基金
安徽省高等学校质量工程项目(2018mooc230)
铜陵职业技术学院自然科研项目(tlpt2019NK004)
铜陵职业技术学院院级研究项目(tlpt2019TG005)。
关键词
叶分量
在线学习
特征提取
图像识别
lobe component
online learning
image feature extraction
image recognition