[1]宋思雨,苗夺谦.基于多粒度空间混乱的细粒度图像分类算法[J].智能系统学报,2022,17(1):144-150.[doi:10.11992/tis.202105040]
SONG Siyu,MIAO Duoqian.Fine-grained image classification algorithm based on multi-granularity regions shuffle[J].CAAI Transactions on Intelligent Systems,2022,17(1):144-150.[doi:10.11992/tis.202105040]
点击复制
SONG Siyu,MIAO Duoqian.Fine-grained image classification algorithm based on multi-granularity regions shuffle[J].CAAI Transactions on Intelligent Systems,2022,17(1):144-150.[doi:10.11992/tis.202105040]
基于多粒度空间混乱的细粒度图像分类算法
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
17
期数:
2022年第1期
页码:
144-150
栏目:
吴文俊人工智能科学技术奖论坛
出版日期:
2022-01-05
- Title:
- Fine-grained image classification algorithm based on multi-granularity regions shuffle
- Keywords:
- fine-grained image classification; regions shuffle; multi-granularity; deep learning; data augmentation; convolutional neural network; weakly-supervised learning; local areas
- 分类号:
- TP183
- 摘要:
- 细粒度图像分类是计算机视觉领域一个具有挑战性的任务，在实际场景中具有很高的应用价值。其中不同子类别的物体整体轮廓差异较小，微小的判别性局部区域是分类的关键。然而，这些重要的局部区域的尺度可能不同, 不能用单一的标准去衡量它们。为了解决这个问题，本文提出了多粒度空间混乱模块来帮助神经网络学习如何寻找到不同尺度的判别性细节。该模块首先将图片划分为不同粒度的局部区域，然后随机打乱并重组构成新的输入图片。经过处理的图片具有区域无关性，从而迫使网络更好地在不同粒度层次下寻找有判别力的局部区域并从中学习特征。在3个广泛使用的细粒度图像分类数据集上的实验证明本文提出的模块可以有效地帮助网络寻找判别性局部区域从而提升了准确率并且网络不需要图片的任何部位标注信息。
- Abstract:
- Fine-grained image classification is a challenging task of computer vision due to the high application value in many reality scenes, having high value in actual application scenarios. The differences in the overall outline of objects from different sub-categories are slight, and the delicate, discriminative local regions are the key to improve the classification accuracy. However, these essential local areas may have different scales, which cannot be measured by a single criterion. Therefore, a multi-granularity regions shuffle module is proposed to help the neural network learn how to find discriminative details for different scales. The module would first divide the image into local areas with different granularity, and then these regions will be shuffled and reorganized to form a new image, which will also be inputted to the network. The irrelevance among regions of the new image forces the network to find discriminative local regions under different granularity and learn from regions. Experimental results of three datasets widely used as benchmarks in fine-grained image classification show that the proposed module can effectively help the network find discriminative local regions and achieve excellent performance with no additional information required to be marked on any part of the image.
- 参考文献/References:
-
[1] 罗建豪, 吴建鑫. 基于深度卷积特征的细粒度图像分类研究综述[J]. 自动化学报, 2017, 43(8): 1306–1318.LUO Jianhao, WU Jianxin. A survey on fine-grained image categorization using deep convolutional features[J]. Acta automatica sinica, 2017, 43(8): 1306–1318.
[2] ZHAO Bo, FENG Jiashi, WU Xiao, et al. A survey on deep learning-based fine-grained object classification and semantic segmentation[J]. International journal of automation and computing, 2017, 14(2): 119–135.
[3] WEI Xiushen, WU Jianxin, CUI Quan. Deep learning for fine-grained image analysis: a survey[EB/OL]. (2019-07-06)[2021-05-26].https://arxiv.org/abs/1907.03069v1.
[4] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521: 436–444.
[5] BENGIO Y, SIMARD P, FRASCONI P. Learning long-term dependencies with gradient descent is difficult[J]. IEEE transactions on neural networks, 1994, 5(2): 157–166.
[6] BENGIO Y, COURVILLE A, VINCENT P. Representation learning: a review and new perspectives[J]. IEEE transactions on pattern analysis and machine intelligence, 2013, 35(8): 1798–1828.
[7] LANG Guangming, MIAO Duoqian, FUJITA H. Three-way group conflict analysis based on pythagorean fuzzy set theory[J]. IEEE transactions on fuzzy systems, 2020, 28(3): 447–461.
[8] YUE X D, CHEN Y F, MIAO D Q, et al. Fuzzy neighborhood covering for three-way classification[J]. Information sciences, 2020, 507: 795–808.
[9] CHEN Yumin, MIAO Duoqian. Granular regression with a gradient descent method[J]. Information sciences, 2020, 537: 247–260.
[10] QIAN Jin, LIU Caihui, MIAO Duoqian, et al. Sequential three-way decisions via multi-granularity[J]. Information sciences, 2020, 507: 606–629.
[11] 王子晔, 苗夺谦, 赵才荣, 等. 基于多粒度特征的行人跟踪检测结合算法[J]. 计算机研究与发展, 2020, 57(5): 996–1002.WANG Ziye, MIAO Duoqian, ZHAO Cairong, et al. A pedestrian tracking algorithm based on multi-granularity feature[J]. Journal of computer research and development, 2020, 57(5): 996–1002.
[12] ZHANG Ning, DONAHUE J, GIRSHICK R, et al. Part-based R-CNNs for fine-grained category detection[C]//Proceedings of the 13th European Conference on Computer Vision. Zurich, Switzerland, 2014: 834–849.
[13] UIJLINGS J R R, VAN DE SANDE K E A, GEVERS T, et al. Selective search for object recognition[J]. International journal of computer vision, 2013, 104(2): 154–171.
[14] BRANSON S, HORN G V, BELONGIE S, et al. Bird species categorization using pose normalized deep convolutional nets[EB/OL]. (2014-06-11)[2021-05-26].https://arxiv.org/abs/1406.2952.
[15] FELZENSZWALB P F, GIRSHICK R B, MCALLESTER D, et al. Object detection with discriminatively trained part-based models[J]. IEEE transactions on pattern analysis and machine intelligence, 2010, 32(9): 1627–1645.
[16] HUANG Shaoli, XU Zhe, TAO Dacheng, et al. Part-Stacked CNN for Fine-Grained Visual Categorization[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, America, 2016: 1173–1182.
[17] LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, America, 2015: 3431–3440.
[18] LIN Di, SHEN Xiaoyong, LU Cewu, et al. Deep LAC: Deep localization, alignment and classification for fine-grained recognition[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, America, 2015: 1666–1674.
[19] WEI Xiushen, XIE Chenwei, WU Jianxin, et al. Mask-CNN: Localizing parts and selecting descriptors for fine-grained bird species categorization[J]. Pattern recognition, 2018, 76: 704–714.
[20] WEI Xiushen, LUO Jianhao, WU Jianxin, et al. Selective convolutional descriptor aggregation for fine-grained image retrieval[J]. IEEE transactions on image processing, 2017, 26(6): 2868–2881.
[21] WANG Dequan, SHEN Zhiqiang, SHAO Jie, et al. Multiple granularity descriptors for fine-grained categorization[C]//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago, Chile, 2015: 2399–2406.
[22] XIAO L, XIA T, WANG J, et al. Fully convolutional attention localization networks: efficient attention localization for fine-grained recognition[EB/OL]. (2016-03-22)[2021-05-26]. https://arxiv.org/abs/1603.06765v1.
[23] LIN T Y, ROYCHOWDHURY A, MAJI S. Bilinear CNN models for fine-grained visual recognition[C]//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago, Chile, 2015: 1449–1457.
[24] LIN T Y, MAJI S. Improved bilinear pooling with CNNs[C]//Proceedings of the British Machine Vision Conference. London, UK, 2017.
[25] CHEN Yue, BAI Yalong, ZHANG Wei, et al. Destruction and construction learning for fine-grained image recognition[C]//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, America, 2019: 5157–5166.
[26] WAH C, BRANSON S, WELINDER P, et al. The caltech-ucsd birds-200-2011 dataset[R]. Pasadena: California Institute of Technology, 2011.
[27] MAJI S, RAHTU E, KANNALA J, et al. Fine-grained visual classification of aircraft[EB/OL]. (2013-06-21)[2021-05-26].https://arxiv.org/abs/1306.5151.
[28] KRAUSE J, STARK M, DENG Jia, et al. 3D object representations for fine-grained categorization[C]//Proceedings of 2013 IEEE International Conference on Computer Vision. Sydney, Australia, 2013: 554–561.
[29] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, America, 2016: 770–778.
[30] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2014-09-04)[2021-05-26]. https://arxiv.org/abs/1409.1556.
- 相似文献/References:
-
[1]刘万军,赵思琪,曲海成,等.结合前景特征增强与区域掩码自注意力的细粒度图像分类[J].智能系统学报,2022,17(6):1134.[doi:10.11992/tis.202109029]
LIU Wanjun,ZHAO Siqi,QU Haicheng,et al.Combining foreground feature reinforcement and region mask self-attention for fine-grained image classification[J].CAAI Transactions on Intelligent Systems,2022,17():1134.[doi:10.11992/tis.202109029]
备注/Memo
收稿日期:2021-05-26。
基金项目:国家自然科学基金项目(61976158,61976160,62076182).
作者简介:宋思雨，硕士研究生，主要研究方向为计算机视觉、深度学习和粒计算;苗夺谦，教授，博士生导师，国际粗糙集学会副理事长、ACM 上海分会学术委员会委员、中国人工智能学会粒计算与知识发现专委会主任，主要研究方向为机器学习、粗糙集、人工智能和粒计算。主持国家自然科学基金项目6项。发表学术论文200余篇， 出版教材及著作21部。
通讯作者:苗夺谦. E-mail：dqmiao@tongji.edu.cn
更新日期/Last Update:
1900-01-01