反向目标干扰的图像数据增强
作者:
基金项目:

辽宁省自然科学基金(20170540426); 辽宁省教育厅重点基金(LJYL049)


Inverse Target Interference for Image Data Augmentation
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [28]
  • |
  • 相似文献 [20]
  • | | |
  • 文章评论
    摘要:

    混合样本数据增强方法只注重模型对于图像所属类别的正向表达, 而忽略图像是否属于某一类别的反向判定. 为了解决描述图像类别方式单一而影响模型性能的问题, 提出一种反向目标干扰的图像数据增强方法. 该方法增加图像背景及目标的多样性, 防止网络模型过拟合. 其次采用反向学习机制, 让网络模型在正确辨别原图像所属类别的同时, 对填充图像不属于该类别的属性进行充分学习, 从而增强网络模型对原图像所属类别辨识的置信度. 最后, 为验证该方法的有效性, 使用不同的网络模型在CIFAR-10、CIFAR-100等5个数据集上进行大量实验. 实验结果表明, 本文方法与其他先进的数据增强方法相比较, 可以显著提高模型在复杂背景下的学习效果和泛化能力.

    Abstract:

    Mixed sample data enhancement methods focus only on the model’s forward representation of the category to which the image belongs while ignoring the reverse determination of whether the image belongs to a specific category. To address the problem of uniquely describing image categories and affecting model performance, this study proposes a method of image data augmentation with inverse target interference. To prevent overfitting of the network model, the method first modifies the original image to increase the diversity of background and target images. Secondly, the idea of reverse learning is adopted to enable the network model to correctly identify the category that the original image belongs to while fully learning the attributes of the populated image that do not belong to that category to increase the confidence of the network model in identifying the category that the original image belongs to. In conclusion, to verify the method’s effectiveness, the study utilizes different network models to perform many experiments on five datasets including CIFAR-10 and CIFAR-100. Experimental results show that compared to other state-of-the-art data augmentation methods, the proposed method can significantly enhance the model’s learning effect and generalization ability in complex settings.

    参考文献
    [1] Shorten C, Khoshgoftaar TM. A survey on image data augmentation for deep learning. Journal of Big Data, 2019, 6(1): 60.
    [2] Su D, Kong H, Qiao YL, et al. Data augmentation for deep learning based semantic segmentation and crop-weed classification in agricultural robotics. Computers and Electronics in Agriculture, 2021, 190: 106418.
    [3] 郭文明, 王腾亿. 类激活映射指导数据增强的细粒度图像分类. 计算机辅助设计与图形学学报, 2021, 33(11): 1698–1704.
    [4] 曾武, 朱恒亮, 邢树礼, 等. 显著性检测引导的图像数据增强方法. 图学学报, 2023, 44(2): 260–270.
    [5] Harris E, Marcu A, Painter M, et al. Fmix: Enhancing mixed sample data augmentation. arXiv:2002.12047, 2021.
    [6] Zhang HY, Cissé M, Dauphin YN, et al. Mixup: Beyond empirical risk minimization. Proceedings of the 6th International Conference on Learning Representations. Vancouver: OpenReview.net, 2018.
    [7] Yun S, Han D, Chun S, et al. CutMix: Regularization strategy to train strong classifiers with localizable features. Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019. 6022–6031.
    [8] Uddin AFMS, Monira MS, Shin W, et al. SaliencyMix: A saliency guided data augmentation strategy for better regularization. arXiv:2006.01791, 2021.
    [9] Bochkovskiy A, Wang CY, Liao HYM. YOLOv4: Optimal speed and accuracy of object detection. arXiv:2004.10934, 2020.
    [10] 马宁, 曹云峰, 王指辉, 等. 基于YOLOv5网络架构的着陆跑道检测算法研究. 激光与光电子学进展, 2022, 59(14): 189–195.
    [11] Koushik J. Understanding convolutional neural networks. arXiv:1605.09081, 2016.
    [12] Kamath U, Liu J, Whitaker J. Deep Learning for NLP and Speech Recognition. Cham: Springer, 2019.
    [13] Cui JF, Zhang XY, Xiong FB, et al. Pathological myopia image recognition strategy based on data augmentation and model fusion. Journal of Healthcare Engineering, 2021, 2021: 5549779.
    [14] He KM, Zhang XY, Ren SQ, et al. Deep residual learning for image recognition. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016. 770–778.
    [15] Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Communications of the ACM, 2017, 60(6): 84–90.
    [16] 马国祥, 严传波, 张志豪, 等. 基于数据增强的CT图像病灶检测方法. 计算机系统应用, 2021, 30(10): 187–194.
    [17] DeVries T, Taylor GW. Improved regularization of convolutional neural networks with cutout. arXiv:1708.04552, 2017.
    [18] Cubuk ED, Zoph B, Mane D, et al. AutoAugment: Learning augmentation policies from data. arXiv:1805.09501, 2019.
    [19] Kim JH, Choo W, Song HO. PuzzleMix: Exploiting saliency and local statistics for optimal mixup. Proceedings of the 37th International Conference on Machine Learning. JMLR.org, 2020. 489.
    [20] Hong MN, Choi J, Kim G. StyleMix: Separating content and style for enhanced data augmentation. Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville: IEEE, 2021. 14857–14865.
    [21] Suzuki T. TeachAugment: Data augmentation optimization using teacher knowledge. Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022. 10894–10904.
    [22] Rahnamayan S, Tizhoosh HR, Salama MMA. Opposition-based differential evolution. In: Chakraborty UK, ed. Advances in Differential Evolution. Berlin: Springer, 2008. 155–177.
    [23] Ventresca M, Tizhoosh HR. Two frameworks for improving gradient-based learning algorithms. In: Tizhoosh HR, Ventresca M, eds. Oppositional Concepts in Computational Intelligence. Berlin: Springer, 2008. 255–284.
    [24] Tizhoosh HR, Sahba F. Quasi-global oppositional fuzzy thresholding. Proceedings of the 2009 IEEE International Conference on Fuzzy Systems. Jeju: IEEE, 2009. 1346–1351.
    [25] Tizhoosh HR, Rahnamayan S. Learning opposites with evolving rules. Proceedings of the 2015 IEEE International Conference on Fuzzy Systems. Istanbul: IEEE, 2015. 1–8.
    [26] Ernoult MM, Normandin F, Moudgil A, et al. Towards scaling difference target propagation by learning backprop targets. Proceedings of the 39th International Conference on Machine Learning. Baltimore: PMLR, 2022. 5968–5987.
    [27] Chen F, Liu YM, Yang J, et al. Multi-objective particle swarm optimization with reverse multi-leaders. Mathematical Biosciences and Engineering, 2023, 20(7): 11732–11762.
    [28] Tizhoosh HR. Opposition-based learning: A new scheme for machine intelligence. Proceedings of the 2005 International Conference on Computational Intelligence for Modelling, Control and Automation and International Conference on Intelligent Agents, Web Technologies and Internet Commerce. IEEE, 2005. 695–701.
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

袁姮,胡月,张晟翀.反向目标干扰的图像数据增强.计算机系统应用,2024,33(6):48-57

复制
分享
文章指标
  • 点击次数:363
  • 下载次数: 917
  • HTML阅读次数: 727
  • 引用次数: 0
历史
  • 收稿日期:2023-12-03
  • 最后修改日期:2023-12-29
  • 在线发布日期: 2024-04-19
文章二维码
您是第11203767位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京海淀区中关村南四街4号 中科院软件园区 7号楼305房间,邮政编码:100190
电话:010-62661041 传真: Email:csa (a) iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号