基于特征优化的低照度小样本目标检测方法

doi:10.3969/j.issn.0255-8297.2026.03.006

应用科学学报 ›› 2026, Vol. 44 ›› Issue (3): 422-436.doi: 10.3969/j.issn.0255-8297.2026.03.006

• 智能信息处理 • 上一篇

基于特征优化的低照度小样本目标检测方法

江泽涛, 金鑫, 冷路, 朱文才

桂林电子科技大学广西图像图形与智能处理重点实验室, 广西桂林 541004

收稿日期:2024-01-28 发布日期:2026-06-23
通信作者: 金鑫，研究方向为图像处理、计算机视觉。E-mail:jx6688110@foxmail.com E-mail:jx6688110@foxmail.com
基金资助:
国家自然科学基金项目(No.62172118);广西自然学科基金重点项目(No.2021GXNSFDA196002);广西图像图形智能处理重点实验项目(No.GIIP2302,No.GIIP2303,No.GIIP2304);桂林电子科技大学研究生教育创新计划项目(No.2023YCXS046)

A Low-Light Few-Shot Object Detection Method Based on Feature Optimization

JIANG Zetao, JIN Xin, LENG Lu, ZHU Wencai

Guangxi Key Laboratory of Image and Graphic Intelligent Processing, Guilin University of Electronic Technology, Guilin 541004, Guangxi, China

Received:2024-01-28 Published:2026-06-23

摘要/Abstract

摘要： 针对低照度目标检测中部分环境存在样本稀缺的问题，提出一种基于特征优化的低照度小样本目标检测方法。该方法设计了降噪Wasserstein自编码器模块和自适应变分特征聚合模块，解决了低照度下图像特征信息不显著问题，突出重要特征；针对小样本学习训练时样本少导致的目标分类混淆问题，设计了类别信息引导检测头模块，减少了分类混淆，提高了检测精度。实验结果表明，相比于所选主流小样本目标检测算法，该方法在低照度数据集上训练后，检测精度平均有9.3%～19.2%的提升，且在正常照度数据集上训练后，该方法相比于目前最先进算法的检测精度平均有3.0%的提升。所提算法对低照度下的小样本目标检测具有积极的意义和良好的应用价值。

关键词: 低照度环境, 小样本目标检测, 自编码器, 特征聚合, 分类混淆

Abstract: To address the scarcity of samples for low-light object detection in certain environments, this paper proposed a low-light few-shot object detection method based on feature optimization. The method designed a denoising Wasserstein autoencoder(DNWAE)module and an adaptive variational feature aggregation(AVFA) module to address the problem of weak image feature information under low-light conditions and enhance important features. To reduce object classification confusion caused by limited training samples in few-shot learning, the paper designed a category information guided detection head(CIGDH) module to improve detection accuracy. Experimental results show that, compared with the selected mainstream few-shot object detection algorithms, this method achieves an average improvement of 9.3%–19.2% in detection accuracy after being trained on low-light datasets. Moreover, after being trained on normal-light datasets, this method achieves an average improvement of 3.0% in detection accuracy compared with the current state-of-the-art algorithm. The proposed algorithm is meaningful and has good application value for few-shot object detection under low-light conditions.

Key words: low-light environment, few-shot object detection, auto encoder, feature aggregation, classification confusion

中图分类号:

TP391.4

江泽涛, 金鑫, 冷路, 朱文才. 基于特征优化的低照度小样本目标检测方法[J]. 应用科学学报, 2026, 44(3): 422-436.

JIANG Zetao, JIN Xin, LENG Lu, ZHU Wencai. A Low-Light Few-Shot Object Detection Method Based on Feature Optimization[J]. Journal of Applied Sciences, 2026, 44(3): 422-436.

参考文献

[1] Yuan Y, Yang W H, Ren W Q, et al. UG2+ Track 2: a collective benchmark effort for evaluating and advancing image understanding in poor visibility environments [DB/OL]. (2019- 04-09) [2024-01-28]. https://arxiv.org/abs/1904.04474.
[2] Neumann L, Karg M, Zhang S, et al. NightOwls: a pedestrians at night dataset [C]//14th Asian Conference on Computer Vision (ACCV 2018), 2019: 691-705.
[3] Loh Y P, Chan C S. Getting to know low-light images with the exclusively dark dataset [J]. Computer Vision and Image Understanding, 2019, 178: 30-42.
[4] Miao Y, Liu F, Hou T, et al. A nighttime vehicle detection method based on YOLO v3[C]//2020 Chinese Automation Congress (CAC), 2020: 6617-6621.
[5] 江泽涛, 翟丰硕, 钱艺, 等. 结合特征增强和多尺度感受野的低照度目标检测[J]. 计算机研究与发展, 2023, 60(4): 903-915. Jiang Z T, Zhai F S, Qian Y, et al. Low illumination object detection combined with feature enhancement and multiscale receptive field [J]. Journal of Computer Research and Development, 2023, 60(4): 903-915. (in Chinese)
[6] Liu W Y, Ren G F, Yu R S, et al. Image-adaptive YOLO for object detection in adverse weather conditions [J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2022, 36(2): 1792-1800.
[7] Yan X P, Chen Z L, Xu A N, et al. Meta R-CNN: towards general solver for instance-level low-shot learning [C]//IEEE/CVF International Conference on Computer Vision (ICCV), 2019: 9576-9585.
[8] Han G X, Ma J W, Huang S Y, et al. Few-shot object detection with fully cross-transformer [C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022: 5311-5320.
[9] Han G X, Huang S Y, Ma J W, et al. Meta faster R-CNN: towards accurate few-shot object detection with attentive feature alignment [C]//AAAI Conference on Artificial Intelligence, 2022, 36(1): 780-789.
[10] Wang X, Huang T E, Darrell T, et al. Frustratingly simple few-shot object detection [C]//37th International Conference on Machine Learning, 2020: 9919-9928.
[11] Wu J, Liu S, Huang D, et al. Multi-scale positive sample refinement for few-shot object detection [C]//European Conference on Computer Vision, 2020: 456-472.
[12] Han J M, Ren Y Q, Ding J, et al. Few-shot object detection via variational feature aggregation [C]//AAAI Conference on Artificial Intelligence, 2023, 37(1): 755-763.
[13] Kingma D P, Welling M. Auto-encoding variational Bayes [DB/OL]. (2013-11-20) [2024-01- 28]. https://arxiv.org/abs/1312.6114.
[14] Tolstikhin I, Bousquet O, Gelly S. Wasserstein auto-encoders [C]//6th International Conference on Learning Representations (ICLR), 2018: 1-16.
[15] Zhu Q L, Bi W, Liu X J, et al. A batch normalized inference network keeps the KL vanishing away [C]//58th Annual Meeting of the Association for Computational Linguistics, 2020: 2636- 2649.
[16] Everingham M, Ali Eslami S M, Van Gool L, et al. The pascal visual object classes challenge: a retrospective [J]. International Journal of Computer Vision, 2015, 111(1): 98-136.
[17] Fan Q, Zhuo W, Tang C K, et al. Few-shot object detection with attention-RPN and multirelation detector [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020: 4012-4021.
[18] Sun B, Li B H, Cai S C, et al. FSCE: few-shot object detection via contrastive proposal encoding [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021: 7348-7358.
[19] Xiao Y, Marlet R. Few-shot object detection and viewpoint estimation for objects in the wild [C]//European Conference on Computer Vision, 2020: 192-210.
[20] Cao Y H, Wang J Q, Jin Y, et al. Few-shot object detection via association and discrimination [C]//35th International Conference on Neural Information Processing Systems, 2021: 16570- 16581.
[21] Zhang L, Zhou S G, Guan J H, et al. Accurate few-shot object detection with support-query mutual guidance and hybrid loss [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021: 14419-14427.
[22] Qiao L M, Zhao Y X, Li Z Y, et al. DeFRCN: decoupled faster R-CNN for few-shot object detection [C]//2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2021: 8661-8670.
[23] Shangguan Z Y, Rostami M. Improved region proposal network for enhanced few-shot object detection [DB/OL]. (2023-08-15) [2024-01-28]. https://arxiv.org/abs/2308.07535.
[24] Wang Z C, Yang B, Yue H N, et al. Fine-grained prototypes distillation for few-shot object detection [J]. AAAI Conference on Artificial Intelligence, 2024, 38(6): 5859-5866.
[25] Gretton A, Borgwardt K M, Rasch M J, et al. A kernel two-sample test [J]. The Journal of Machine Learning Research, 2012, 13(1): 723-773.

基于特征优化的低照度小样本目标检测方法

A Low-Light Few-Shot Object Detection Method Based on Feature Optimization

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

[1]	黄逸凡, 陆小锋, 孙军, 唐嘉吕, 刘学锋. 基于分割分类多任务学习的轻量化脑血管造影质量评估模型[J]. 应用科学学报, 2025, 43(4): 672-683.
[2]	王莹笑, 杨彦红, 谭云峰. 面向视频的人脸特征计算方法[J]. 应用科学学报, 2025, 43(1): 137-153.
[3]	李国瑞, 田丽, 崔浩, 陈浩波. 一种基于自编码器的无线传感网数据收集方案[J]. 应用科学学报, 2018, 36(3): 411-419.