基于卷积神经网络的多肉植物细粒度图像分类

doi:10.12066/j.issn.1007-2861.2029

上海大学学报(自然科学版) ›› 2020, Vol. 26 ›› Issue (2): 283-291.doi: 10.12066/j.issn.1007-2861.2029

基于卷积神经网络的多肉植物细粒度图像分类

黄嘉宝¹, 朱永华², 周霁婷¹(), 高文靖¹

^1. 上海大学上海电影学院, 上海 200072
^2. 上海大学计算机工程与科学学院, 上海 200444

收稿日期:2018-03-09 出版日期:2020-04-30 发布日期:2020-04-29
通讯作者: 周霁婷 E-mail:zjting@163.com
基金资助:
十三五规划重点研究发展计划资助项目(2017YFD0400101)

Fine-grained image classification of succulents with convolutions

HUANG Jiabao¹, ZHU Yonghua², ZHOU Jiting¹(), GAO Wenjing¹

^1. Shanghai Film Academy, Shanghai University, Shanghai 200072, China
^2. School of Computer Engineering and Science, Shanghai University, Shanghai 200444, China

Received:2018-03-09 Online:2020-04-30 Published:2020-04-29
Contact: ZHOU Jiting E-mail:zjting@163.com

摘要/Abstract

摘要：

多肉植物分类是植物栽培管理中的一项重要任务,通常需使用大型数据集和领域独有的特性. 由于没有现成的多肉植物数据集,需收集大量的图片自制数据集. 研究了多肉植物的细粒度图像分类.为了识别不同视角、背景、光效和成长阶段的多肉植物,对卷积神经网络 AlexNet 和 GoogLeNet 的最后三层进行微调,对原创数据集进行了强监督分类和弱监督分类的测试、训练. 实验结果表明,微调 GoogLeNet 的强监督分类达到了最佳效果, 精准率为 96.7${\%}$.

关键词: 细粒度图像分类, 强监督分类, 弱监督分类, 卷积神经网络, AlexNet, GoogLeNet, 微调

Abstract:

Succulents recognition is an important task in plants identification and management, and domain-specific features and large datasets are used. Due to the absence of an existing succulents' dataset, a large number of pictures need to be collected. This paper addresses the issue of fine-grained image classification of succulent plants. Both supervised and weakly supervised fine-tuning of AlexNet and GoogLeNet training on original dataset were imple mented, trained and tested for the task of identifying the class of succulents from various viewpoints, backgrounds, light effects and growth stages. Results showed that supervised fine-tuning GoogLeNet improved the performance, and the accuracy rate could be as high as 96.7${\%}$.

Key words: fine-grained image classification, supervised classification, weakly supervised classification, convolutional neural networks (CNNs), AlexNet, GoogLeNet, fine-tuning

中图分类号:

TP183

黄嘉宝, 朱永华, 周霁婷, 高文靖. 基于卷积神经网络的多肉植物细粒度图像分类[J]. 上海大学学报(自然科学版), 2020, 26(2): 283-291.

HUANG Jiabao, ZHU Yonghua, ZHOU Jiting, GAO Wenjing. Fine-grained image classification of succulents with convolutions[J]. Journal of Shanghai University（Natural Science Edition）, 2020, 26(2): 283-291.

图/表 12

表 1

图 1

图 2

表 2

表 3

图 3

表 4

图 4

图 5

图 6

表 5

表 6

参考文献 22

[1]	Krizhevsky A, Sutskever I, Hinton G. ImageNet classification with deep convolutional neural networks[C]// International Conference on Neural Information Processing. 2012: 1097-1105.
[2]	Pan S J, Yang Q. A survey on transfer learning[C]// IEEE Transactions on Knowledge and Data Engineering. 2010: 1345-1359.
[3]	Donahue J, Jia Y Q, Vinyals O, et al. DeCAF: a deep convolutional activation feature for generic visual recognition[C]// Proceedings of the 31st International Conference on Machine Learning. 2014: 647-655.
[4]	Branson S, Horn G V, Belongie S, et al. Bird species categorization using pose normalized deep convolutional nets[C]// British Machine Vision Conference. 2014.
[5]	Huang S L, Xu Z, Tao D C, et al. Part-stacked CNN for fine-grained visual categoriza-tion[C]// Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2016: 1173-1182.
[6]	Zhang X, Xiong H, Zhou W, et al. Picking deep filter responses for fine-grained image recognition[C]// Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2016.
[7]	Zhang H, Xu T, Elhoseiny M, et al. SPDA-CNN: unifying semantic part detection and abstraction for fine-grained recognition[C]// Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2016.
[8]	Lam M, Mahasseni B, Todorovic S. Fine-grained recognition as HSNet search for informative image parts[C]// Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2017.
[9]	He X T, Peng Y X, Zhao J J . Fast fine-grained image classification via weakly supervised discriminative localization[C]// 2019 IEEE International Conference on Multimedia and Expo (ICME). 2019.
[10]	Peng Y X, He X T, Zhao J J. Object-part attention driven discriminative localization for fine-grained image classification[C]// Proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2017.
[11]	Angelova A, Zhu S. Efficient object detection and segmentation for fine-grained recognition[C]// IEEE Transactions on Knowledge and Data Engineering. 2013.
[12]	Cui Y, Zhou F, Lin Y Q, et al. Fine-grained categorization and dataset bootstrapping using deep metric learning with humans in the loop[C]// Proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2015.
[13]	Ghazi M, Yanikoglu B, Aptoula E . Plant identification using deep neural networks via optimization of transfer learning parameters[J]. Neurocomputing, 2017,235:228-235.
[14]	Zhou F, Lin Y. Fine-grained image classification by exploring bipartite graph labels[C]// Proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2016.
[15]	Yang L, Luo P, Chen C, et al. A large-scale car dataset for fine-grained categorization and verification[C]// Proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2015.
[16]	Sochor J, Herout A, Havel J. BoxCars: 3D boxes as CNN input for improved fine-grained vehicle recognition[C]// Proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2016.
[17]	Wang Y, Choi J, Morariu V, et al. Mining discriminative triplets of patches for fine-grained classification[C]// Proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2016.
[18]	Liu D, Wang Y . Monza: image classification of vehicle make and model using convolutional neural networks and transfer learning[M]. San Fransisco: Stanford University, 2017.
[19]	Maji S, Rahtu E, Kannala J , et al. Fine-grained visual classification of aircraft [J]. Computer Science, 2013. arXiv:1306.5151.
[20]	Krause J, Jin H, Yang J, et al. Fine-grained recognition without part annota-tions[C]// Proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2015.
[21]	Christian S. Going deeper with convolutions[C]// Proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2014.
[22]	Jia Y Q, Shelhamer E, Donahue J. Caffe: convolutional architecture for fast feature embedding[C]// Proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2014.

基于卷积神经网络的多肉植物细粒度图像分类

Fine-grained image classification of succulents with convolutions

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 22

相关文章 3

编辑推荐

Metrics

本文评价

[1]	周敬一, 郭燕, 丁友东. 基于深度学习的中文影评情感分析[J]. 上海大学学报(自然科学版), 2018, 24(5): 703-712.
[2]	冉鹏, 王灵, 李昕, 刘鹏伟. 改进Softmax分类器的深度卷积神经网络及其在人脸识别中的应用[J]. 上海大学学报(自然科学版), 2018, 24(3): 352-366.
[3]	刘惠彬, 陈强, 吴飞, 赵毅. 基于卷积神经网络的固定群体中目标人物分类[J]. 上海大学学报(自然科学版), 2017, 23(6): 874-.