计算机与数字工程2023，Vol.51Issue(10)：2395-2399,2424,6.DOI:10.3969/j.issn.1672-9722.2023.10.034
基于图像表示和稀疏表示的图像分类
Image Classification Based on Image Representation and Sparse Representation
摘要
Abstract
In order to better realize the image classification task,this paper proposes a new image classification algorithm.This algorithm can better retain the large-scale information of the original image,effectively use the global characteristics of the orig-inal image,and reduce the difference of the same object in different images,which significantly improves the accuracy of image classification.The algorithm first generates a virtual image by a new image representation algorithm,and then uses image classifica-tion algorithm to obtain the corresponding classification error for the original image and the virtual image respectively.Finally,the fi-nal classification error is obtained by a simple and efficient error fusion scheme.Experimental results show that the algorithm effec-tively reduces the error rate of image classification.关键词
图像表示/图像分类/稀疏表示/全局特征Key words
image representation/image classification/sparse representation/global image features分类
信息技术与安全科学引用本文复制引用
潘承昌,张永军,王泽伟,刘竣文..基于图像表示和稀疏表示的图像分类[J].计算机与数字工程,2023,51(10):2395-2399,2424,6.基金项目
国家自然科学基金项目(编号:62062023) （编号:62062023）
贵州省教育厅创新群体研究项目(编号:黔教合KY字[2021]022) （编号:黔教合KY字[2021]022）
贵州省研究生科研基金项目(编号:YJSCXJH[2020]53,YJSCXJH[2020]189)资助. （编号:YJSCXJH[2020]53,YJSCXJH[2020]189）
