An Integrated Active Deep Learning Approach for Image Classification from Unlabeled Data with Minimal Supervision

被引:2
作者
Abdelwahab, Amira [1 ,2 ]
Afifi, Ahmed [3 ,4 ]
Salama, Mohamed [5 ]
Kim, Byung-Gyu
机构
[1] King Faisal Univ, Coll Comp Sci & Informat Technol, Dept Informat Syst, POB 400, Al Hasa 31982, Saudi Arabia
[2] Menoufia Univ, Fac Comp & Informat, Dept Informat Syst, Shibin Al Kawm 32511, Menoufia, Egypt
[3] King Faisal Univ, Coll Comp Sci & Informat Technol, Dept Comp Sci, POB 400, Al Hasa 31982, Saudi Arabia
[4] Menoufia Univ, Fac Comp & Informat, Dept Informat Technol, Shibin Al Kawm 32511, Menoufia, Egypt
[5] Future Acad, Higher Inst Specif Studies, Dept Management Informat Syst, Heliopolis POB 11757, Cairo, Egypt
关键词
active learning; unlabeled data classification; query strategies; deep learning; image classification; annotation costs;
D O I
10.3390/electronics13010169
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The integration of active learning (AL) and deep learning (DL) presents a promising avenue for enhancing the efficiency and performance of deep learning classifiers. This article introduces an approach that seamlessly integrates AL principles into the training process of DL models to build robust image classifiers. The proposed approach employs a unique methodology to select high-confidence unlabeled data points for immediate labeling, reducing the need for human annotation and minimizing annotation costs. Specifically, by combining uncertainty sampling with the pseudo-labeling of confident data, the proposed approach expands the training set efficiently. The proposed approach uses a hybrid active deep learning model that selects the most informative data points that need labeling based on an uncertainty measure. Then, it iteratively retrains a deep neural network classifier on the newly labeled samples. The model achieves high accuracy with fewer manually labeled samples than traditional supervised deep learning by selecting the most informative samples for labeling and retraining in a loop. Experiments on various image classification datasets demonstrate that the proposed model outperforms conventional approaches in terms of classification accuracy and reduced human annotation requirements. The proposed model achieved accuracy of 98.9% and 99.3% for the Cross-Age Celebrity and Caltech Image datasets compared to the conventional approach, which achieved 92.3% and 74.3%, respectively. In summary, this work presents a promising unified active deep learning approach to minimize the human effort in manually labeling data while maximizing classification accuracy by strategically labeling only the most valuable samples for the model.
引用
收藏
页数:14
相关论文
共 36 条
[1]  
Biswas S., 2020, Neural Netw, V133, P19
[2]  
Cacciarelli D, 2023, Arxiv, DOI arXiv:2302.08893
[3]   A meta-framework for multi-label active learning based on deep reinforcement learning [J].
Chen, Shuyue ;
Wang, Ran ;
Lu, Jian .
NEURAL NETWORKS, 2023, 162 :258-270
[4]  
Chen X., 2021, Proc. AAAI Conf. Artif. Intell, V35, P3980
[5]   Application of EfficientNet-B0 and GRU-based deep learning on classifying the colposcopy diagnosis of precancerous cervical lesions [J].
Chen, Xiaoyue ;
Pu, Xiaowen ;
Chen, Zhirou ;
Li, Lanzhen ;
Zhao, Kong-Nan ;
Liu, Haichun ;
Zhu, Haiyan .
CANCER MEDICINE, 2023, 12 (07) :8690-8699
[6]   Attention-Based Active Learning Framework for Segmentation of Breast Cancer in Mammograms [J].
Fu, Xianjun ;
Cao, Hao ;
Hu, Hexuan ;
Lian, Bobo ;
Wang, Yansong ;
Huang, Qian ;
Wu, Yirui .
APPLIED SCIENCES-BASEL, 2023, 13 (02)
[7]  
Gal Y., 2016, P INT C LEARNING REP
[8]   Seismic Stratigraphic Interpretation Based on Deep Active Learning [J].
Gu, Xiaofeng ;
Lu, Wenkai ;
Ao, Yile ;
Li, Yinshuo ;
Song, Cao .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2023, 61
[9]   Active learning for efficient analysis of high-throughput nanopore data [J].
Guan, Xiaoyu ;
Li, Zhongnian ;
Zhou, Yueying ;
Shao, Wei ;
Zhang, Daoqiang .
BIOINFORMATICS, 2023, 39 (01)
[10]   Informativeness-guided active learning for deep learning-based facade defects detection [J].
Guo, Jingjing ;
Wang, Qian ;
Su, Shu ;
Li, Yiting .
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING, 2023, 38 (17) :2408-2425