《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (2): 592-598.DOI: 10.11772/j.issn.1001-9081.2021061109

• 多媒体计算与计算机仿真 • 上一篇    

基于教与学优化的可变卷积自编码器的医学图像分类方法

李薇1,2, 樊瑶驰1(), 江巧永1, 王磊2, 徐庆征3   

  1. 1.西安理工大学 计算机科学与工程学院, 西安 710048
    2.陕西省网络计算与安全技术重点实验室(西安理工大学), 西安 710048
    3.国防科技大学 信息通信学院, 西安 710106
  • 收稿日期:2021-06-28 修回日期:2021-07-14 接受日期:2021-07-15 发布日期:2022-02-21 出版日期:2022-02-10
  • 通讯作者: 樊瑶驰
  • 作者简介:李薇(1973—),女,黑龙江哈尔滨人,副教授,博士,CCF会员,主要研究方向:演化优化、数据挖掘;
    樊瑶驰(1996—),男,陕西西安人,硕士研究生,CCF学生会员,主要研究方向:演化优化、机器学习;
    江巧永(1987—),男,浙江台州人,讲师,博士,CCF会员,主要研究方向:演化优化;
    王磊(1972—),男,陕西白水人,教授,博士,CCF高级会员,主要研究方向:演化优化、大数据;
    徐庆征(1980—),男,山东临清人,副教授,博士,CCF高级会员,主要研究方向:演化优化、最优化理论与应用。

Variable convolutional autoencoder method based on teaching-learning-based optimization for medical image classification

Wei LI1,2, Yaochi FAN1(), Qiaoyong JIANG1, Lei WANG2, Qingzheng XU3   

  1. 1.School of Computer Science and Engineering,Xi’an University of Technology,Xi’an Shaanxi 710048,China
    2.Shaanxi Key Laboratory for Network Computing and Security Technology (Xi’an University of Technology),Xi’an Shaanxi 710048,China
    3.College of Information and Communication,National University of Defense Technology,Xi’an Shaanxi 710106,China
  • Received:2021-06-28 Revised:2021-07-14 Accepted:2021-07-15 Online:2022-02-21 Published:2022-02-10
  • Contact: Yaochi FAN
  • About author:LI Wei, born in 1973, Ph. D., associate professor. Her research interests include evolutionary optimization, data mining.
    FAN Yaochi, born in 1996, M. S. candidate. His research interests include evolutionary optimization, machine learning.
    JIANG Qiaoyong, born in 1987, Ph. D., lecturer. His research interests include evolutionary optimization.
    WANG Lei, born in 1972, Ph. D., professor. His research interests include evolutionary optimization, big data.
    XU Qingzheng, born in 1980, Ph. D., associate professor. His research interests include evolutionary optimization, optimization theory and application.

摘要:

针对传统手工方法优化卷积神经网络(CNN)参数时存在耗时长、不准确,以及参数设置影响算法性能等问题,提出一种基于教与学优化(TLBO)的可变卷积自编码器(CAE)算法。该算法设计了可变长度的个体编码策略,从而快速构建CAE结构,并堆叠CAE为一个CNN;此外,充分利用优秀个体的结构信息来引导算法朝着更有希望的区域搜索,从而提高算法性能。实验结果表明,所提算法在解决医学图像分类问题时,分类精度达到89.84%,高于传统CNN和同类型神经网络。该算法通过优化CAE结构和堆叠CNN解决医学图像分类问题,有效提高了医学图像分类性能。

关键词: 卷积自编码器, 卷积神经网络, 教与学优化, 演化算法, 医学图像

Abstract:

In order to solve the problems such as high time cost, inaccuracy and influence of parameter setting on algorithm performance when optimizing parameters of Convolutional Neural Network (CNN) by traditional manual methods, a variable Convolutional AutoEncoder (CAE) method based on Teaching-Learning-Based Optimization (TLBO) was proposed. In the algorithm, a variable-length individual encoding strategy was designed to quickly construct the CAE structure, and stack CAEs to a CNN. In addition, the excellent individual structure information was fully utilized to guide the algorithm to search the regions with more possibility, thereby improving the algorithm performance. Experimental results show that the classification accuracy of the proposed algorithm achieves 89.84% when solving medical image classification problems, which is higher than those of traditional CNN and similar neural networks. The proposed algorithm solves the medical image classification problems by optimizing the CAE structure and stacking CNN, and effectively improves the classification accuracy of medical image classification.

Key words: Convolutional AutoEncoder (CAE), Convolutional Neural Network (CNN), Teaching-Learning-Based Optimization (TLBO), Evolutionary Algorithm (EA), medical image

中图分类号: