Image Augmentation for Classifying Facial Expression Images by Using Deep Neural Network Pre-trained with Object Image Database

被引：13
|
作者：
Shima, Yoshihiro [1 ]
Omori, Yuki [1 ]
机构：
[1] Meisei Univ, Sch Sci & Engn, 2-1-1 Hodokubo, Hino, Tokyo 1918506, Japan
来源：
PROCEEDINGS OF ICRCA 2018: 2018 THE 3RD INTERNATIONAL CONFERENCE ON ROBOTICS, CONTROL AND AUTOMATION / ICRMV 2018: 2018 THE 3RD INTERNATIONAL CONFERENCE ON ROBOTICS AND MACHINE VISION | 2018年
关键词：
Computer vision; facial expression; feature extraction; neural networks; support vector machine; deep learning; RECOGNITION;
D O I：
10.1145/3265639.3265664
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
Image augmentation of automatic facial expression classification is proposed on the basis of a combination of a deep neural network and a support vector machine. A neural network pre-trained with a large-scale object image database is used as a feature extractor for facial images. The accuracy of system performance is evaluated using the database "ATR Facial Expression Image Database (DB99)." By using image augmentation, an average recognition rate of 97.92% was obtained, which was a 9.84 percentage point improvement compared with that without augmentation. The experimental results showed the effectiveness of our scheme.
引用
收藏
页码：140 / 146
页数：7
相关论文
共 50 条
  • [1] Image Hashing by Pre-Trained Deep Neural Network
    Li Pingyuan
    Zhang Dan
    Yuan Xiaoguang
    Jiang Suiping
    2022 ASIA CONFERENCE ON ALGORITHMS, COMPUTING AND MACHINE LEARNING (CACML 2022), 2022, : 468 - 471
  • [2] Image Augmentation for Object Image Classification Based On Combination of Pre-Trained CNN and SVM
    Shima, Yoshihiro
    2ND INTERNATIONAL CONFERENCE ON MACHINE VISION AND INFORMATION TECHNOLOGY (CMVIT 2018), 2018, 1004
  • [3] Classifying for a Mixture of Object Images and Character Patterns by Using CNN Pre-trained for Large-scale Object Image Dataset
    Shima, Yoshihiro
    Nakashima, Yumi
    Yasuda, Michio
    PROCEEDINGS OF THE 2018 13TH IEEE CONFERENCE ON INDUSTRIAL ELECTRONICS AND APPLICATIONS (ICIEA 2018), 2018, : 2360 - 2365
  • [4] On Pre-trained Image Features and Synthetic Images for Deep Learning
    Hinterstoisser, Stefan
    Lepetit, Vincent
    Wohlhart, Paul
    Konolige, Kurt
    COMPUTER VISION - ECCV 2018 WORKSHOPS, PT I, 2019, 11129 : 682 - 697
  • [5] SAR Image Despeckling Using Pre-trained Convolutional Neural Network Models
    Yang, Xiangli
    Denis, Loic
    Tupin, Florence
    Yang, Wen
    2019 JOINT URBAN REMOTE SENSING EVENT (JURSE), 2019,
  • [6] Red Green Blue Depth Image Classification Using Pre-Trained Deep Convolutional Neural Network
    Kumar, N.
    Kaur, N.
    Gupta, D.
    PATTERN RECOGNITION AND IMAGE ANALYSIS, 2020, 30 (03) : 382 - 390
  • [7] Red Green Blue Depth Image Classification Using Pre-Trained Deep Convolutional Neural Network
    N. Kumar
    N. Kaur
    D. Gupta
    Pattern Recognition and Image Analysis, 2020, 30 : 382 - 390
  • [8] Fast Learning for Accurate Object Recognition Using a Pre-trained Deep Neural Network
    Lobato-Rios, Victor
    Tenorio-Gonzalez, Ana C.
    Morales, Eduardo F.
    ADVANCES IN SOFT COMPUTING, MICAI 2017, PT I, 2018, 10632 : 41 - 53
  • [9] Pre-trained SAM as data augmentation for image segmentation
    Wu, Junjun
    Rao, Yunbo
    Zeng, Shaoning
    Zhang, Bob
    CAAI TRANSACTIONS ON INTELLIGENCE TECHNOLOGY, 2025, 10 (01) : 268 - 282
  • [10] Medical Image Classification: A Comparison of Deep Pre-trained Neural Networks
    Alebiosu, David Olayemi
    Muhammad, Fermi Pasha
    2019 17TH IEEE STUDENT CONFERENCE ON RESEARCH AND DEVELOPMENT (SCORED), 2019, : 306 - 310
12345