Deep one-class probability learning for end-to-end image classification

被引:0
|
作者
Liu, Jia [1 ]
Zhang, Wenhua [1 ]
Liu, Fang [1 ]
Yang, Jingxiang [1 ]
Xiao, Liang [1 ]
机构
[1] Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, Nanjing 210094, Peoples R China
关键词
One-class learning; Image classification; Probabilistic model; Deep neural network; SUPPORT;
D O I
10.1016/j.neunet.2025.107201
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
One-class learning has many application potentials in novelty, anomaly, and outlier detection systems. It aims to distinguish both positive and negative samples with a model trained via only positive samples or one-class annotated samples. With the difficulty in training an end-to-end classification network, existing methods usually make decisions indirectly. To fully exploit the learning capability of a deep network, in this paper, we propose to design a deep end-to-end binary image classifier based on convolutional neural network with input of image and output of classification result. Without negative training samples, we establish a probabilistic model driven by an energy to learn the distribution of positive samples. The energy is proposed based on the output of the network which subtly models the deep discriminations into statistics. During optimization, to overcome the difficulty of distribution estimation, we propose a novel particle swarm optimization algorithm based sampling method. Compared with existing methods, the proposed method is able to directly output classification results without additional thresholding or estimating operations. Moreover, the deep network is directly optimized via the probabilistic model which results in better adaptation of positive distribution and classification task. Experiments demonstrate the effectiveness and state-of-the-art performance of the proposed method.
引用
收藏
页数:14
相关论文
共 50 条
  • [21] End-to-End Deep Learning Model for Corn Leaf Disease Classification
    Amin, Hassan
    Darwish, Ashraf
    Hassanien, Aboul Ella
    Soliman, Mona
    IEEE ACCESS, 2022, 10 : 31103 - 31115
  • [22] Deep Learning Methods for Bug Bite Classification: An End-to-End System
    Ilijoski, Bojan
    Dineva, Katarina Trojachanec
    Ribarski, Biljana Tojtovska
    Petrov, Petar
    Mladenovska, Teodora
    Trajanoska, Milena
    Gjorshoska, Ivana
    Lameski, Petre
    APPLIED SCIENCES-BASEL, 2023, 13 (08):
  • [23] Classification of ALS Point Clouds Using End-to-End Deep Learning
    Winiwarter, Lukas
    Mandiburger, Gottfried
    Schmohl, Stefan
    Pfeifer, Norbert
    PFG-JOURNAL OF PHOTOGRAMMETRY REMOTE SENSING AND GEOINFORMATION SCIENCE, 2019, 87 (03): : 75 - 90
  • [24] Exploring End-to-end Deep Learning Applications for Event Classification at CMS
    Andrews, Michael
    Paulini, Manfred
    Gleyzer, Sergei
    Poczos, Barnabas
    23RD INTERNATIONAL CONFERENCE ON COMPUTING IN HIGH ENERGY AND NUCLEAR PHYSICS (CHEP 2018), 2019, 214
  • [25] Classification of ALS Point Clouds Using End-to-End Deep Learning
    Lukas Winiwarter
    Gottfried Mandlburger
    Stefan Schmohl
    Norbert Pfeifer
    PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science, 2019, 87 : 75 - 90
  • [26] New Results in End-to-end Image and Video Compression by Deep Learning
    Ozsoy, Gokberk
    Yilmaz, Melih
    Kirmemis, Ogun
    Tekalp, A. Murat
    2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU), 2020,
  • [27] End-to-End Deep ROI Image Compression
    Akutsu, Hiroaki
    Naruko, Takahiro
    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2020, E103D (05): : 1031 - 1038
  • [28] One-Class Risk Estimation for One-Class Hyperspectral Image Classification
    Zhao, Hengwei
    Zhong, Yanfei
    Wang, Xinyu
    Shu, Hong
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2023, 61
  • [29] End-to-end encrypted network traffic classification method based on deep learning
    Tian Shiming
    Gong Feixiang
    Mo Shuang
    Li Meng
    Wu Wenrui
    Xiao Ding
    TheJournalofChinaUniversitiesofPostsandTelecommunications, 2020, 27 (03) : 21 - 30
  • [30] End-to-end deep learning classification of vocal pathology using stacked vowels
    Liu, George S.
    Hodges, Jordan M.
    Yu, Jingzhi
    Sung, C. Kwang
    Erickson-DiRenzo, Elizabeth
    Doyle, Philip C.
    LARYNGOSCOPE INVESTIGATIVE OTOLARYNGOLOGY, 2023, 8 (05): : 1312 - 1318