基于邻域融合和特征增强的小样本细粒度图像分类

doi:10.19678/j.issn.1000-3428.0070136

摘要/Abstract

摘要：

目前, 细粒度图像分类任务面临着标注困难、样本数量稀缺以及类别差异微小等挑战。为了应对这些问题, 提出一种基于邻域融合和特征增强的小样本细粒度图像分类方法。首先, 利用离散余弦变换(DCT)和通道注意力机制分别捕获图像的全局信息和局部信息, 并将这2种特征在通道维度上进行拼接, 这种结合空间域和频率域的特征提取方法不仅增强了样本特征的多样化, 还提高了模型的泛化能力; 其次, 引入特征增强模块计算查询样本与支持类原型之间的相关性, 生成自适应权重, 以指导查询信息, 补充支持样本图像的细致学习, 这一过程有效地捕捉了同类别图像之间的差异, 同时抑制不同类别图像的局部相似性; 最后, 使用双相似性度量模块衡量支持类原型与待分类样本图像之间的相关分数, 实现更精准的图像分类。实验结果表明, 在Mini-ImageNet、CUB-200-2011、Stanford Dogs和Stanford Cars 4个公开数据集的5-shot任务中, 该方法的准确率分别达到了79.22%、87.47%、79.23%和83.71%, 相较于对比方法性能更优。

关键词: 小样本学习, 细粒度图像分类, 频域学习, 特征增强, 注意力机制, 度量学习

Abstract:

Currently fine-grained image classification faces challenges such as labeling difficulties, scarce sample numbers, and subtle category differences. To address these issues, a few-shot fine-grained image classification method based on neighborhood fusion and feature enhancement is proposed. First, the Discrete Cosine Transform (DCT) and channel attention mechanisms are used to capture global and local information from images, respectively. These features are then concatenated along the channel dimensions. This method of combining spatial- and frequency-domain feature extraction enhances the diversity of sample features and improves model generalization. Second, a feature enhancement module is introduced to compute the correlation between query samples and support class prototypes, generating adaptive weights to guide query information to complement the detailed learning of support sample images. This process effectively captures the differences between images of the same class and suppresses local similarities between different classes. Finally, a dual-similarity measurement module assesses the correlation scores between the support class prototypes and the images to be classified, improving the accuracy of classification performance. The experimental results show that this method achieves accuracies of 79.22%, 87.47%, 79.23%, and 83.71% on the 5-shot tasks in the Mini-ImageNet, CUB-200-2011, Stanford Dogs, and Stanford Cars datasets, respectively, outperforming comparative methods.

Key words: few-shot learning, fine-grained image classification, frequency domain learning, feature enhancement, attention mechanism, metric learning

文浪, 苟光磊, 白瑞峰, 缪宛谕. 基于邻域融合和特征增强的小样本细粒度图像分类[J]. 计算机工程, 2026, 52(2): 158-166.

WEN Lang, GOU Guanglei, BAI Ruifeng, MIAO Wanyu. Few-shot Fine-grained Image Classification Based on Neighborhood Fusion and Feature Enhancement[J]. Computer Engineering, 2026, 52(2): 158-166.

https://www.ecice06.com/CN/Y2026/V52/I2/158

图/表 13

图1 细粒度图像对比

Fig.1 Fine-grained image comparison

图2 本文网络框架

Fig.2 The framework of the network in this paper

图3 邻域融合模块

Fig.3 Neighborhood fusion module

图4 特征增强模块

Fig.4 Feature enhancement module

图5 损失函数变化曲线

Fig.5 Loss function variation curve

图6 可视化效果对比

Fig.6 Visual effect comparison

参考文献 35

1	XIE D Z, PANG C, WU G H, et al. Feature channel adaptive enhancement for fine-grained visual classification[EB/OL]. [2024-05-05]. https://link.springer.com/chapter/10.1007/978-3-031-47665-5_16.
2	CAO S Y, WANG W, ZHANG J, et al. A few-shot fine-grained image classification method leveraging global and local structures. International Journal of Machine Learning and Cybernetics, 2022, 13(8): 2273- 2281. doi: 10.1007/s13042-022-01522-w
3	ZHANG Y. A fine-grained image classification and detection method based on convolutional neural network fused with attention mechanism. Computational Intelligence and Neuroscience, 2022, 2022, 2974960.
4	HUANG H X, ZHANG J J, ZHANG J, et al. Compare more nuanced: pairwise alignment bilinear network for few-shot fine-grained learning[C]//Proceedings of the IEEE International Conference on Multimedia and Expo (ICME). Washington D.C., USA: IEEE Press, 2019: 91-96.
5	ZHENG Z J, FENG X, YU H Q, et al. BDLA: bi-directional local alignment for few-shot learning. Applied Intelligence, 2023, 53(1): 769- 785. doi: 10.1007/s10489-022-03479-3
6	贺小箭, 林金福. 融合弱监督目标定位的细粒度小样本学习. 中国图象图形学报, 2022, 27(7): 2226- 2239.
	HE X J, LIN J F. Weakly-supervised object localization based fine-grained few-shot learning. Journal of Image and Graphics, 2022, 27(7): 2226- 2239.
7	GE W F, LIN X R, YU Y Z. Weakly supervised complementary parts models for fine-grained image classification from the bottom up[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D.C., USA: IEEE Press, 2020: 3029-3038.
8	白尚旺, 王梦瑶, 胡静, 等. 多区域注意力的细粒度图像分类网络. 计算机工程, 2024, 50(1): 271- 278. doi: 10.19678/j.issn.1000-3428.0066426
	BAI S W, WANG M Y, HU J, et al. Multi-region attention network for fine-grained image classification. Computer Engineering, 2024, 50(1): 271- 278. doi: 10.19678/j.issn.1000-3428.0066426
9	QIAN L L, YU T, YANG J Y. Multi-scale feature fusion of covariance pooling networks for fine-grained visual recognition. Sensors, 2023, 23(8): 3970. doi: 10.3390/s23083970
10	李小雨, 罗娜. 基于迁移类内变化增强数据的小样本学习方法. 计算机工程, 2025, 51(9): 242- 251. doi: 10.19678/j.issn.1000-3428.0069754
	LI X Y, LUO N. Few-shot learning method with augmentation data based on transferring intra-class variations. Computer Engineering, 2025, 51(9): 242- 251. doi: 10.19678/j.issn.1000-3428.0069754
11	FINN C, ABBEEL P, LEVINE S. Model-agnostic meta-learning for fast adaptation of deep networks[C]// Proceedings of the International Conference on Machine Learning. Washington D.C., USA: IEEE Press, 2017: 12-24.
12	SNELL J, SWERSKY K, ZEMEL R. Prototypical networks for few-shot learning. Advances in Neural Information Processing Systems, 2017, 30, 35- 42.
13	YAN L L, LI F Z, ZHANG L, et al. Discriminant space metric network for few-shot image classification. Applied Intelligence, 2023, 53(14): 17444- 17459. doi: 10.1007/s10489-022-04413-3
14	WEI X S, WANG P, LIU L Q, et al. Piecewise classifier mappings: learning fine-grained learners for novel categories with few examples. IEEE Transactions on Image Processing, 2019, 28(12): 6116- 6125. doi: 10.1109/TIP.2019.2924811
15	SONG Q, ZHOU S, XU L. Learning more discriminative local descriptors for few-shot learning[EB/OL]. [2024-05-05]. https://arxiv.org/abs/2305.08721.
16	MENG X X, WANG X W, YIN S L, et al. Few-shot image classification algorithm based on attention mechanism and weight fusion. Journal of Engineering and Applied Science, 2023, 70(1): 14. doi: 10.1186/s44147-023-00186-9
17	YANG Y C, SOATTO S. FDA: Fourier domain adaptation for semantic segmentation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D.C., USA: IEEE Press, 2020: 4084-4094.
18	LIN H, TSE R, TANG S K, et al. Few-shot learning for plant-disease recognition in the frequency domain. Plants, 2022, 11(21): 2814. doi: 10.3390/plants11212814
19	CHENG H, YANG S Y, ZHOU J T, et al. Frequency guidance matters in few-shot learning[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). Washington D.C., USA: IEEE Press, 2024: 11780-11790.
20	QIN Z Q, ZHANG P Y, WU F, et al. FcaNet: frequency channel attention networks[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). Washington D.C., USA: IEEE Press, 2022: 763-772.
21	LIU H F, PENG P, CHEN T, et al. FECANet: boosting few-shot semantic segmentation with feature-enhanced context-aware network. IEEE Transactions on Multimedia, 2023, 25, 8580- 8592. doi: 10.1109/TMM.2023.3238521
22	LI X X, WU J J, SUN Z, et al. BSNet: bi-similarity network for few-shot fine-grained image classification. IEEE Transactions on Image Processing, 2021, 30, 1318- 1331. doi: 10.1109/TIP.2020.3043128
23	LIAO J J, LEWIS J W. A note on concordance correlation coefficient. PDA Journal of Pharmaceutical Science and Technology, 2000, 54(1): 23- 26.
24	DENG J, DONG W, SOCHER R, et al. ImageNet: a large-scale hierarchical image database[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Washington D.C., USA: IEEE Press, 2009: 248-255.
25	SONG H O, XIANG Y, JEGELKA S, et al. Deep metric learning via lifted structured feature embedding[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Washington D.C., USA: IEEE Press, 2016: 4004-4012.
26	KHOSLA A, JAYADEVAPRAKASH N, YAO B, et al. Novel dataset for fine-grained image categorization: Stanford Dogs[EB/OL]. [2024-05-05]. https://people.csail.mit.edu/khosla/papers/fgvc2011.pdf.
27	KRAUSE J, STARK M, JIA D, et al. 3D object representations for fine-grained categorization[C]//Proceedings of the IEEE International Conference on Computer Vision Workshops. Washington D.C., USA: IEEE Press, 2014: 554-561.
28	VINYALS O, BLUNDELL C, LILLICRAP T, et al. Matching networks for one shot learning[C]//Proceedings of the 30th International Conference on Neural Information Processing Systems. Washington D.C., USA: IEEE Press, 2016: 3637-3645.
29	SUNG F, YANG Y X, ZHANG L, et al. Learning to compare: relation network for few-shot learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D.C., USA: IEEE Press, 2018: 1199-1208.
30	LI W B, WANG L, XU J L, et al. Revisiting local descriptor based image-to-class measure for few-shot learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D.C., USA: IEEE Press, 2020: 7253-7260.
31	XUE Z, DUAN L, LI W, et al. Region comparison network for interpretable few-shot image classification[EB/OL]. [2024-05-05]. https://arxiv.org/abs/2009.03558.
32	OCHOA R G, MENDEZ R M, GONZALEZ-ZAPATA J, et al. Enforcing class separability in metric learning via two novel distance-based loss functions for few-shot image classification[EB/OL]. [2024-05-05]. https://arxiv.org/abs/2305.09062.
33	YANG Y J, FENG Y X, ZHU L, et al. Feature fusion network based on few-shot fine-grained classification. Frontiers in Neurorobotics, 2023, 17, 1301192. doi: 10.3389/fnbot.2023.1301192
34	JIANG Z H, KANG B Y, ZHOU K Q, et al. Few-shot classification via adaptive attention[EB/OL]. [2024-05-05]. https://arxiv.org/abs/2008.02465.
35	XU Q Y, SU J, WANG Y, et al. Few-shot learning based on double pooling squeeze and excitation attention. Electronics, 2023, 12(1): 27.

[1]	李健浪, 吴新电, 陈灵, 阳波, 唐文胜. 基于4D毫米波雷达与视觉融合的三维目标检测算法[J]. 计算机工程, 2026, 52(2): 299-310.
[2]	张信佳, 王芳. 基于多层次特征融合和注意力机制的无人机图像小目标检测算法[J]. 计算机工程, 2026, 52(2): 148-157.
[3]	刘畅, 梁冰雪, 田荣坤, 秦玉华. 基于多特征融合和混合神经网络的医疗健康问题分类[J]. 计算机工程, 2026, 52(2): 342-355.
[4]	宋朝琦, 刘颖, 何敬鲁, 李大湘. 基于显著位置交互Transformer的小样本图像分类方法[J]. 计算机工程, 2026, 52(2): 167-176.
[5]	王庆荣, 郝福乐, 朱昌锋, 王俊杰. 基于多特征融合的车辆轨迹预测研究[J]. 计算机工程, 2026, 52(2): 331-341.
[6]	许晓阳, 魏伟, 高重阳. 基于改进YOLOv7-tiny的红外船舶目标检测[J]. 计算机工程, 2026, 52(2): 209-220.
[7]	薛阳, 秦瑶, 张舒翔. 基于双重图注意力网络生成子图的图神经协同推荐[J]. 计算机工程, 2026, 52(2): 89-100.
[8]	但崇鸿, 韦洪雷, 何舟, 吴贯锋. SRMpose: 一种多尺度特征提取的关键点检测算法[J]. 计算机工程, 2026, 52(2): 136-147.
[9]	黎东丰, 陈雨人, 余博. 基于多层次特征融合的路面裂缝检测方法[J]. 计算机工程, 2026, 52(1): 154-165.
[10]	王珺, 李昆仑, 张伊菲, 朱其振, 刘磊, 王帅. 基于位置和语义分离注意力机制的轻量视频目标跟踪算法[J]. 计算机工程, 2026, 52(1): 228-241.
[11]	黄金贵, 刘朋, 唐文胜. MMD-YOLOv7:黑暗条件下车辆检测方法[J]. 计算机工程, 2025, 51(9): 340-349.
[12]	符家成, 田瑾, 张玉金, 方志军. 结合前置三元组集的知识图谱推荐[J]. 计算机工程, 2025, 51(9): 101-109.
[13]	李小雨, 罗娜. 基于迁移类内变化增强数据的小样本学习方法[J]. 计算机工程, 2025, 51(9): 242-251.
[14]	马淦, 谷雨, 彭冬亮. 结合改进YOLOv5s和动态数据增强的海面舰船检测[J]. 计算机工程, 2025, 51(9): 294-305.
[15]	陈彦如, 刘珂良, 冉茂亮. 基于深度强化学习的外卖即时配送实时优化[J]. 计算机工程, 2025, 51(9): 328-339.

选择文件类型/文献管理软件名称

选择包含的内容