Semantic segmentation of blue-green algae based on deep generative adversarial net

doi:10.11772/j.issn.1001-9081.2017122872

Abstract

Abstract: Concerning the problem of insufficient accuracy of the traditional image segmentation algorithm in segmentation of blue-green alga images, a new network structure named Deep Generative Adversarial Net (DGAN) based on Deep Neural Network (DNN) and Generative Adversarial Net (GAN) was proposed. Firstly, based on Fully Convolutional neural Network (FCN), a 12-layer FCN was constructed as the Generater (G), which was used to study the distribution of data and generate the segmentation result of blue-green alga images (Fake). Secondly, a 5-layer Convolutional Neural Network (CNN) was constructed as the Discriminator (D), which was used to distinguish the segmentation result generated by the generated network (Fake) and the true segmentation result with manual annotation (Label), G tried to generate Fake and deceive D, D tried to find out Fake and punish G. Finally, through the adversarial training of two networks, a better segmentation result was obtained because Fake generated by G could cheat D. The training and test results on image sets with 3075 blue-green alga images show that, the proposed DGAN is far ahead of the iterative threshold segmentation algorithm in precision, recall and F₁ score, which are increased by more than 4 percentage points than other DNN algorithms such as FCNNet (SHELHAMER E, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651) and Deeplab (CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Semantic image segmentation with deep convolutional nets and fully connected CRFs. Computer Science, 2014(4):357-361). The proposed DGAN has obtained more accurate segmentation results. In the aspect of segmentation speed, the DGAN needs 0.63 s per image, which is slightly slower than the traditional FCNNet with 0.46 s, but much faster than Deeplab with 1.31 s. The balanced segmentation accuracy and speed of DGAN can provide a feasible technical scheme for image-based semantic segmentation of blue-green algae.

Key words: Generative Adversarial Net (GAN), Deep Neural Network (DNN), Fully Convolutional neural Network (FCN), blue-green algae, semantic segmentation

摘要： 针对传统图像分割算法分割蓝藻图像准确率不足的问题，提出了一种基于深度神经网络（DNN）和生成式对抗网络（GAN）思想的网络结构，称为深度生成式对抗网络（DGAN）。首先，在传统全卷积神经网络（FCN）的基础上构建了一个12层的FCN作为生成网络（G），用于学习分布规律，生成蓝藻图像的分割结果（Fake）；然后，设计了一个5层的卷积神经网络（CNN）作为判别网络（D），用于区分生成网络生成的分割结果（Fake）和手工标注的真实分割结果（Label），G试图生成Fake并蒙骗D，D试图找出Fake并惩罚G；最后，通过两个网络的对抗式训练，G生成的Fake可以蒙骗D，从而获得了更好的分割结果。在3075张蓝藻图像集上的训练和测试结果表明，DGAN在精确率、召回率及F₁分数等指标上均大幅领先基于迭代的阈值分割算法；相比FCNNet （SHELHAMER E，LONG J，DARRELL T.Fully convolutional networks for semantic segmentation.IEEE Transactions on Pattern Analysis and Machine Intelligence，2017，39（4）：640-651）、Deeplab （CHEN L C，PAPANDREOU G，KOKKINOS I，et al.Semantic image segmentation with deep convolutional nets and fully connected CRFs.Computer Science，2014（4）：357-361）等其他基于DNN的方法也提升了超过4个百分点，取得了更精准的分割结果。分割速度上，DGAN的0.63 s略慢于FCNNet的0.46 s，但远快于Deeplab的1.31 s。DGAN均衡的分割准确率和分割速度为基于图像的蓝藻语义分割提供了可行的技术方案。

关键词: 生成式对抗网络, 深度神经网络, 全卷积神经网络, 蓝藻, 语义分割

CLC Number:

TP391.4

YANG Shuo, CHEN Lifang, SHI Yu, MAO Yiming. Semantic segmentation of blue-green algae based on deep generative adversarial net[J]. Journal of Computer Applications, 2018, 38(6): 1554-1561.

杨朔, 陈丽芳, 石瑀, 毛一鸣. 基于深度生成式对抗网络的蓝藻语义分割[J]. 计算机应用, 2018, 38(6): 1554-1561.

References

[1] 林怡,潘琛,陈映鹰,等.基于遥感影像光谱分析的蓝藻水华识别方法[J].同济大学学报(自然科学版),2011,39(8):1247-1252.(LIN Y, PAN C, CHEN Y Y, et al. Recognition of cyanobacteria bloom based on spectral analysis of remote sensing imagery[J]. Journal of Tongji University (Natural Science), 2011, 39(8):1247-1252.)
[2] 陈云,戴锦芳.基于遥感数据的太湖蓝藻水华信息识别方法[J].湖泊科学,2008,20(2):179-183.(CHEN Y, DAI J F. Extraction methods of cyanobacteria bloom in Lake Taihu based on RS data[J]. Journal of Lake Sciences, 2008, 20(2):179-183.)
[3] 李亚春,谢小萍,朱小莉,等.结合卫星遥感技术的太湖蓝藻水华形成温度特征分析[J].湖泊科学,2016,28(6):1256-1264.(LI Y C, XIE X P, ZHU X L, et al. Applying remote sensing techniques in analysis of temperature features causing cyanobacteria bloom in Lake Taihu[J]. Journal of Lake Sciences, 2016, 28(6):1256-1264.)
[4] SHELHAMER E, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651.
[5] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Semantic image segmentation with deep convolutional nets and fully connected CRFs[EB/OL].[2017-10-26]. http://www.ee.bgu.ac.il/~rrtammy/DNN/reading/SemanticYuille.pdf.
[6] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[C]//Proceedings of the 2014 International Conference on Neural Information Processing Systems. Cambridge, MA:MIT Press, 2014:2672-2680.
[7] 焦李成,杨淑媛,刘芳,等.神经网络七十年:回顾与展望[J].计算机学报,2016,39(8):1697-1716.(JIAO L C, YANG S Y, LIU F, et al. Seventy years beyond neural networks:retrospect and prospect[J]. Chinese Journal of Computers, 2016, 39(8):1697-1716.)
[8] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11):2278-2324.
[9] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2017-10-26]. http://x-algo.cn/wp-content/uploads/2017/01/VERY-DEEP-CONVOLUTIONAL-NETWORK-SFOR-LARGE-SCALE-IMAGE-RECOGNITION.pdf.
[10] SZEGEDY C, LIU W, JIA Y Q, et al. Going deeper with convolutions[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2015:1-9.
[11] SRIVASTAVA N, HINTON G, KRIZHEVSKY A, et al. Dropout:a simple way to prevent neural networks from overfitting[J]. Journal of Machine Learning Research, 2014, 15(1):1929-1958.
[12] IOFFE S, SZEGEDY C. Batch normalization:accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 201532nd International Conference on International Conference on Machine Learning. Cambridge, MA:MIT Press, 2015:448-456.
[13] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2016:770-778.
[14] RONNEBERGER O, FISCHER P, BROX T. U-Net:convolutional networks for biomedical image segmentation[C]//MICCAI 2015:Proceedings of the 2015 International Conference on Medical Image Computing and Computer-Assisted Intervention, LNCS 9351. Cham:Springer, 2015:234-241.
[15] KINGMA D P, BA J L. Adam:a method for stochastic optimization[EB/OL].[2017-10-26]. http://yeolab.weebly.com/uploads/2/5/5/0/25509700/a_method_for_stochastic_optimization_.pdf.