计算机科学与探索 ›› 2024, Vol. 18 ›› Issue (4): 990-1000.DOI: 10.3778/j.issn.1673-9418.2302015

• 图形·图像 • 上一篇    下一篇

采用特征图增强原型的小样本图像分类方法

许华杰,梁书伟   

  1. 1. 广西大学 计算机与电子信息学院,南宁 530004
    2. 广西多媒体通信与网络技术重点实验室,南宁 530004
    3. 广西高校并行分布与智能计算重点实验室,南宁 530004
    4. 广西智能数字服务工程技术研究中心,南宁 530004
  • 出版日期:2024-04-01 发布日期:2024-04-01

Few-Shot Image Classification Method with Feature Maps Enhancement Prototype

XU Huajie, LIANG Shuwei   

  1. 1. College of Computer and Electronic Information, Guangxi University, Nanning 530004, China
    2. Guangxi Key Laboratory of Multimedia Communications and Network Technology, Nanning 530004, China
    3. Guangxi Key Laboratory of Parallel, Distributed and Intelligent Computing, Nanning 530004, China
    4. Guangxi Intelligent Digital Services Research Center of Engineering Technology, Nanning 530004, China
  • Online:2024-04-01 Published:2024-04-01

摘要: 在基于度量学习的小样本图像分类方法中,由于标注样本的稀缺,仅用支持集样本得到的类原型往往难以代表整个类别的真实分布;同时,同类样本间也可能在多个方面存在较大差异,较大的类内差异可能使样本特征偏离类别中心。针对上述可能严重影响图像分类性能的问题,提出一种采用特征图增强原型的小样本图像分类方法(FMEP)。首先,用余弦相似度从查询集样本特征图中选择部分相似特征加入类原型中,得到更具代表性的特征图增强原型;其次,对相似的查询集样本特征进行聚合,缓解类内差异大导致的问题,使同类样本的特征分布更接近;最后,用在特征空间中与真实类别分布都更接近的特征图增强原型和聚合查询特征进行相似度比较得到更优的分类结果。所提方法在MiniImageNet、TieredImageNet、CUB-200和CIFAR-FS等常用的小样本图像分类数据集上进行了实验,结果表明所提方法获得了比基线模型更优的分类性能,同时也优于同类型的小样本图像分类方法。

关键词: 小样本学习, 图像分类, 度量学习, 特征图增强原型, 余弦相似度

Abstract: Due to the scarcity of labeled samples, the class prototype obtained by support set samples is difficult to represent the real distribution of the whole class in metric-based few-shot image classification methods. Meanwhile, samples of the same class may also have large difference in many aspects and the large intra-class bias may make the sample features deviate from the class center. Aiming at the above problems that may seriously affect the performance, a few-shot image classification method with feature maps enhancement prototype (FMEP) is proposed. Firstly, this paper selects some similar features of the query set sample feature maps with cosine similarity and adds them to class prototypes to obtain more representative prototypes. Secondly, this paper aggregates similar features of the query set to alleviate the problem caused by large intra-class bias and makes features distribution of the same class closer. Finally, this paper compares enhanced prototypes and aggregated features which are both closer to real distribution to get better results. The proposed method is tested on four commonly used few-shot classification datasets, namely MiniImageNet, TieredImageNet, CUB-200 and CIFAR-FS. The results show that the proposed method can not only improve the performance of the baseline model, but also obtain better performance compared with the same type of methods.

Key words: few-shot learning, image classification, metric learning, feature maps enhancement prototype, cosine similarity