###
计算机系统应用英文版:2021,30(1):174-179
本文二维码信息
码上扫一扫!
基于矢量表示组合的字典学习性能分析
(1.电子科技大学 信息与通信工程学院, 成都 610054;2.贵州大学 贵州省公共大数据重点实验室, 贵阳 550025;3.贵州大学 计算机科学与技术学院, 贵阳 550025;4.哈尔滨工业大学 深圳研究生院, 深圳 518055)
Dictionary Learning Performance Analysis Based on Combination of Vector Representations
(1.School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 610054, China;2.Guizhou Provincial Key Laboratory of Public Big Data, Guizhou University, Guiyang, 550025, China;3.College of Computer Science & Technology, Guizhou University, Guiyang 550025, China;4.Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen 518055, China)
摘要
图/表
参考文献
相似文献
本文已被:浏览 48次   下载 77
Received:June 03, 2020    Revised:June 30, 2020
中文摘要: 在字典学习算法中, 使用图像的多矢量表示相比单一矢量表示, 可以获得分类精度更高且更具有鲁棒性的分类模型. 本文中我们采用多种矢量表示的组合以及合理的加权对数和方案, 来提升字典算法的性能. 通过在公共人脸数据集上进行实验, 验证了我们的方法应用于字典学习具有更高的准确度和鲁棒性. 充分挖掘和利用表示多样性可以获得被观察对象的各种潜在外观以及图像高分类精度.
Abstract:In the dictionary learning algorithms, the model by the multi-vector representation can obtain better classification performance and more robustness than that by the single vector representation. In this study, we use the combined representation fused multiple vector representations and reasonable weighted logarithms sum schemes to improve the performance of the dictionary algorithm. Experiments on public face datasets verify that dictionary learning algorithms applied with proposed method has higher accuracy and robustness. It illustrates that the various potential appearances of observed objects generated by fully mining and utilizing the diversity of representations are beneficial to improve the performance of images classification.
文章编号:     中图分类号:    文献标志码:
基金项目:贵州省科技计划项目(黔科合重大专项字[2018]3001); 贵州省公共大数据重点实验室开放课题(2018BDKFJJ001)
引用文本:
焦健雄,孙利雷,徐勇.基于矢量表示组合的字典学习性能分析.计算机系统应用,2021,30(1):174-179
JIAO Jian-Xiong,SUN Li-Lei,XU Yong.Dictionary Learning Performance Analysis Based on Combination of Vector Representations.COMPUTER SYSTEMS APPLICATIONS,2021,30(1):174-179