摘要
针对目前数字图像识别准确率不高的问题,重点研究了基于小波空间特征谱熵的图像特征提取方法。该方法利用小波变换前后能量不变的原理,构造小波能量模式矩阵,对该矩阵进行奇异值分解,并求取了奇异值的特征谱熵作为图像的特征。实验结果表明,本文提出的图像特征提取方法能够获得很高的图像正确识别率,证明了该方法的有效性和实用性。
The current recognition accuracy of digital image is lower.To overcome this problem,the method of image feature extraction based on wavelet space feature spectrum entropy is proposed.According to the principle that the energy is equal before and after the wavelet transform,the matrix of wavelet energy model is constructed.Then,the singular value decomposition of the matrix is performed.Finally,the feature spectrum entropy of the singular value is taken as the image feature.Experimental results show that the proposed image feature extraction method can obtain higher image recognition accuracy,and the effectiveness and practicality are proved.
出处
《吉林大学学报（工学版）》
EI
CAS
CSCD
北大核心
2015年第6期1994-1998,共5页
Journal of Jilin University:Engineering and Technology Edition
基金
国家自然科学基金项目(61272351)
福建省自然科学基金项目(2013J01017)
关键词
图像处理
数字图像
特征提取
小波变换
特征谱熵
image processing
digital image
feature extraction
wavelet transform
feature spectrum entropy
