帮助 关于我们

返回检索结果

基于主动学习的图像分类技术:现状与未来
Active Learning-Based Image Classification Technology: Status and Future

查看参考文献126篇

刘颖 1,2   庞羽良 1 *   张伟东 1,2   李大湘 1,2   许志杰 2,3  
文摘 图像分类作为计算机视觉领域中的重要研究方向之一,应用领域非常广泛.基于深度学习的图像分类技术取得的成功,依赖大量的已标注数据,然而数据的标注成本往往是昂贵的.主动学习作为一种机器学习方法,旨在以尽可能少的高质量标注数据达到期望的模型性能,缓解监督学习任务中存在的标注成本高、标注信息难以大量获取的问题.主动学习图像分类算法根据样本选择策略,从未标记样本数据集合中选择出信息量丰富,对分类模型训练贡献更高的样本进行标注,以更新已标注训练数据池,如此循环直至满足给定的停止条件或模型标注预算耗尽.本文对近年来提出的主动学习图像分类算法进行了详细综述,并根据所用样本数据处理及模型优化方案,将现有算法分为三类:基于数据增强的算法,包括利用图像增广来扩充训练数据,或者根据图像特征插值后的差异性来选择高质量的训练数据;基于数据分布信息的算法,根据数据分布的特点来优化样本选择策略;优化模型预测的算法,包括优化获取和利用深度模型预测信息的方法、基于生成对抗网络和强化学习来优化预测模型的结构,以及基于Transformer结构提升模型预测性能,以确保模型预测结果的可靠性.此外,本文还对各类主动学习图像分类算法下的重要学术工作进行了实验对比,并对各算法在不同规模数据集上的性能和适应性进行了分析.另外,本文探讨了主动学习图像分类技术所面临的挑战,并指出了未来研究的方向.
其他语种文摘 As one of the important research directions in the field of computer vision, image classification has a wide range of applications. The success of deep learning-based image classification techniques depends on a large amount of annotated data. However, the cost of data annotation is often expensive. Active learning is a machine learning method that aims to achieve the expected model performance with as few high-quality annotated data as possible, and it can alleviate the problem of high annotation costs and difficulty in obtaining a large amount of annotation information in supervised learning tasks. Based on a sample selection strategy, active learning for image classification selects samples from the unlabeled dataset which are informative and thus contribute more to the training of the classification model, in order to update the annotated training data pool. This process is repeated until a given stopping condition is met or the model annotation budget is exhausted. This paper provides a comprehensive survey of the active learning image classification algorithms published in recent years. According the strategies applied in sample data processing and model structure optimization, existing algorithms are classified into three categories: algorithms based on data augmentation, including those using image augmentation to expand the scale of training data or using the differences in image feature interpolation to select high-quality training data; algorithms based on data distribution information, which optimize sample selection strategies based on the characteristics of data distribution; algorithms for optimizing model predictions, including methods for optimizing the acquisition and utilization of deep model prediction information, improving the predictive model structure through the use of generative adversarial networks and reinforcement learning, as well as enhancing model prediction performance based on the Transformer architecture to ensure the reliability of model predictions. In addition, this paper also conducts experimental comparisons on important academic work under various types of active learning image classification algorithms, and analyzes the performance and adaptability of each algorithm on datasets of different scales. Furthermore, this paper discusses the challenges faced by active learning image classification technology and points out future research directions.
来源 电子学报 ,2023,51(10):2960-2984 【核心库】
DOI 10.12263/DZXB.20230397
关键词 图像分类 ; 主动学习 ; 数据增强 ; 数据分布 ; 模型预测信息 ; 模型结构优化
地址

1. 西安邮电大学图像与信息处理研究所, 陕西, 西安, 710121  

2. 无线通信与信息处理技术国际联合研究中心, 无线通信与信息处理技术国际联合研究中心, 陕西, 西安, 710121  

3. 英国哈德斯菲尔德大学, 西约克郡, HD13DH

语种 中文
文献类型 研究性论文
ISSN 0372-2112
学科 自动化技术、计算机技术
基金 国家自然科学基金青年项目
文献收藏号 CSCD:7607879

参考文献 共 126 共7页

1.  Lecun Y. Deep learning. Nature,2015,521(7553):436-444 CSCD被引 3555 次    
2.  Gu J X. Recent advances in convolutional neural networks. Pattern Recognition,2018,77:354-377 CSCD被引 267 次    
3.  El-Sawy A. CNN for handwritten Arabic digits recognition based on LeNet-5. International Conference on Advanced Intelligent Systems and Informatics,2017:566-575 CSCD被引 1 次    
4.  Ballester P. On the performance of GoogLeNet and AlexNet applied to sketches. Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence,2016:1124-1128 CSCD被引 3 次    
5.  Krizhevsky A. ImageNet classification with deep convolutional neural networks. Communications of the ACM,2017,60(6):84-90 CSCD被引 3458 次    
6.  Simonyan K. Very deep convolutional networks for large-scale image recognition,2014 CSCD被引 547 次    
7.  He K M. Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2016:770-778 CSCD被引 673 次    
8.  Kothawade S. CLINICAL: Targeted active learning for imbalanced medical image classification. Workshop on Medical Image Learning with Limited and Noisy Data,2022:119-129 CSCD被引 1 次    
9.  Tuia D. A survey of active learning algorithms for supervised remote sensing image classification. IEEE Journal of Selected Topics in Signal Processing,2011,5(3):606-617 CSCD被引 20 次    
10.  Settles B. Active learning literature survey,2010 CSCD被引 17 次    
11.  Beluch W H. The power of ensembles for active learning in image classification. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition,2018:9368-9377 CSCD被引 2 次    
12.  Mackowiak R. CEREALS-cost-effective REgion-based active learning for semantic segmentation,2018 CSCD被引 2 次    
13.  Brust C A. Active learning for deep object detection,2018 CSCD被引 1 次    
14.  Sener O. Active learning for convolutional neural networks: A core-set approach,2017 CSCD被引 2 次    
15.  Gal Y. Deep Bayesian active learning with image data. Proceedings of the 34th International Conference on Machine Learning,2017:1183-1192 CSCD被引 8 次    
16.  Sun L. Active learning for image classification: A deep reinforcement learning approach. 2019 2nd China Symposium on Cognitive Computing and Hybrid Intelligence (CCHI),2019:71-76 CSCD被引 1 次    
17.  Zhu J J. Generative adversarial active learning,2017 CSCD被引 2 次    
18.  Mahapatra D. Efficient active learning for image classification and segmentation using a sample selection and conditional generative adversarial network. International Conference on Medical Image Computing and Computer-Assisted Intervention,2018:580-588 CSCD被引 1 次    
19.  Bengar J Z. Class-balanced active learning for image classification. 2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV),2022:3707-3716 CSCD被引 1 次    
20.  Liu P. A survey on active deep learning: From model driven to data driven. ACM Computing Surveys,54(10s):221 CSCD被引 2 次    
引证文献 1 篇

1 姜文涛 门控机制的图像分类网络 电子学报,2024,52(7):2393-2406
CSCD被引 0 次

显示所有1篇文献

论文科学数据集
PlumX Metrics
相关文献

 作者相关
 关键词相关
 参考文献相关

版权所有 ©2008 中国科学院文献情报中心 制作维护：中国科学院文献情报中心
地址：北京中关村北四环西路33号 邮政编码：100190 联系电话：(010)82627496 E-mail:cscd@mail.las.ac.cn 京ICP备05002861号-4 | 京公网安备11010802043238号