浏览全部资源
扫码关注微信
导出
分享
收藏
专辑
纸质出版日期： 2021-03-16 ，
录用日期： 2020-04-24
移动端阅览
徐知宇, 周艺, 王世新, 王丽涛, 王振庆. 面向GF-2遥感影像的U-Net城市绿地分类[J]. 中国图象图形学报, 2021,26(3):700-713.
Zhiyu Xu, Yi Zhou, Shixin Wang, Litao Wang, Zhenqing Wang. U-Net for urban green space classification in Gaofen-2 remote sensing images[J]. Journal of Image and Graphics, 2021,26(3):700-713.
目的
2
高分2号卫星(GF-2)是首颗民用高空间分辨率光学卫星，具有亚米级高空间分辨率与宽覆盖结合的显著特点，为城市绿地信息提取等多领域提供了重要的数据支撑。本文利用GF-2卫星多光谱遥感影像，将一种改进的U-Net卷积神经网络首次应用于城市绿地分类，提出一种面向高分遥感影像的城市绿地自动分类提取技术。
方法
2
先针对小样本训练集容易产生的过拟合问题对U-Net网络进行改进，添加批标准化(batch normalization，BN)和dropout层获得U-Net+模型；再采用随机裁剪和随机数据增强的方式扩充数据集，使得在充分利用影像信息的同时保证样本随机性，增强模型稳定性。
结果
2
将U-Net+模型与最大似然法(maximum likelihood estimation，MLE)、神经网络(neural networks，NNs)和支持向量机(support vector machine，SVM)3种传统分类方法以及U-Net、SegNet和DeepLabv3+这3种深度学习语义分割模型进行分类结果精度对比。改进后的U-Net+模型能有效防止过拟合，模型总体分类精度比改进前提高了1.06%。基于改进的U-Net+模型的城市绿地总体分类精度为92.73%，平均F
1
分数为91.85%。各分类方法按照总体分类精度从大到小依次为U-Net+(92.73%)、U-Net (91.67%)、SegNet (88.98%)、DeepLabv3+(87.41%)、SVM (81.32%)、NNs (79.92%)和MLE (77.21%)。深度学习城市绿地分类方法能充分挖掘数据的光谱、纹理及潜在特征信息，有效降低分类过程中产生的"椒盐噪声"，具有较好的样本容错能力，比传统遥感分类方法更适用于城市绿地信息提取。
结论
2
改进后的U-Net+卷积神经网络模型能够有效提升高分遥感影像城市绿地自动分类提取精度，为城市绿地分类提供了一种新的智能解译方法。
Objective
2
High-precision monitoring of the spatial distribution of urban green space has important social
economic
and ecological benefits for optimizing the spatial structure of such space
maintaining urban ecological balance
and developing green city construction. As the first civilian optical satellite with high spatial resolution
Gaofen2 (GF-2) exhibits the remarkable characteristics of sub-meter high spatial resolution and wide coverage. GF2 provides important data support to multiple fields
such as urban environmental monitoring and urban green space information extraction. However
traditional classification methods still encounter many problems. For example
training a method to be an effective classifier for massive data is difficult
and the accuracy of classification results is generally low. The use of massive high-resolution remote sensing images to achieve large-scale rapid and accurate urban green space distribution extraction is an urgent task for urban planning managers. With the rapid development of deep learning technology
full convolutional networks (FCN) provide novel creative possibilities for semantic segmentation and realize pixel-level classification of images in the field of deep learning for the first time. Inspired by the U-Net network structure
we applied an improved U-Net to urban green space classification for the first time and proposed an automatic classification technique for urban green space by using high-resolution remote sensing images.
Method
2
First
we improved the U-Net model to obtain the U-Net+ model. The main structure of U-Net+ is composed of an encoder and a decoder that can achieve end-to-end training. The encoding channel realizes the multi-scale feature recognition of an image through four-time maximum pooling
and the decoding channel restores the position and detailed information of an image through upsampling. The network uses skip connection to realize the fusion of feature information with the same scale at different levels
overcoming accuracy loss caused by upsampling. In addition
we improved the model by adding batch normalization (BN) after each layer of network convolution operation
effectively regulating the input of the network layer and improving model training speed and network generalization capability. To solve the overfitting problem
which is easily produced by the limited sample training set
we added the dropout layer with a 50% probability of dropping neurons after the convolution operation of the fourth and fifth layers of the network. Second
deep learning requires a large amount of label data related to the classification objectives for training. However
existing open-source datasets cannot meet the requirements of the urban green space classification task. Manually establishing an urban green space tag dataset is necessary. We selected three typical urban green space sample areas in Beijing (urban parks
residential areas
and golf courses) as study areas. By combining GF-2 images and Google Earth remote sensing images in summer and winter
we drew all types of urban green space in the study areas through visual interpretation by using ArcGIS. The visual interpretation results are corrected with actual field investigation. Third
random cropping and data augmentation techniques are adopted to expand the dataset
ensuring the randomness of the samples and enhancing the stability of the model while fully utilizing image information. We adopt the Adam optimizer with an initial learning rate of 0.000 1.
Result
2
1) The overall classification accuracy of the U-Net+ model is improved by 1.06% compared with that of the original U-Net. After 40 training epochs
the accuracy of the U-Net+ model reaches a high level
and the loss function realizes rapid convergence. The U-Net+ model effectively prevents overfitting and improves generalization capability. 2) To verify the effectiveness of our method
the classification accuracy of the U-Net+ results is compared with those of three traditional classification methods
namely
maximum likelihood estimation (MLE)
neural networks (NNs)
and support vector machine (SVM)
and three semantic segmentation models
i.e.
U-Net
SegNet
and DeepLabv3+. Among the seven classification methods
the U-Net+ model achieves the highest overall classification accuracy for urban green space. The seven classification methods are arranged in order of classification accuracy from large to small: U-Net+ (92.73%)
>
U-Net (91.67%)
>
SegNet (88.98%)
>
DeepLabv3+ (87.41%)
>
SVM (81.32%)
>
NNs (79.92%)
>
MLE (77.21%). 3) In the three types of urban green space
evergreen trees have the highest classification accuracy (
F
1
=93.65%)
followed by grassland (
F
1
=92.55%) and deciduous trees (
F
1
=86.55%). 4) Deep learning exhibits strong fault-tolerant capability for training samples. By training and learning a large number of label data
it can effectively reduce the impact of errors and improve recognition capability
making it more suitable for urban green space information extraction than traditional remote sensing classification methods.
Conclusion
2
Deep learning urban green space classification methods can fully mine the spectral
textural
and potential feature information of data. Meanwhile
the U-Net+ model proposed in this study can also effectively reduce the salt-and-pepper noise the classification process and realize high-precision pixel-level classification of urban green space. The improved U-Net+ can effectively improve the accuracy of automatic classification of urban green space in high-resolution remote sensing images and provide a new intelligent interpretation method for urban green space classification in the future.
城市绿地卷积神经网络U-Net高分遥感语义分割
urban green spaceconvolutional neural network (CNN)U-Nethigh-resolution remote sensingsemantic segmentation
Badrinarayanan V, Kendall A and Cipolla R. 2017. SegNet: a deep convolutional encoder-decoder architecture for image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(12): 2481-2495[DOI:10.1109/TPAMI.2016.2644615]
Beijing Municipal Institute of city Planning and Design. 2010. Planning of green space system in Beijing[EB/OL].[2020-02-19].http://yllhj.beijing.gov.cn/zwgk/ghxx/gh/201911/P0201911295-02960993686.pdfhttp://yllhj.beijing.gov.cn/zwgk/ghxx/gh/201911/P0201911295-02960993686.pdf
北京市城市规划设计研究院. 2010. 北京市绿地系统规划[EB/OL]. [2020-02-19].http://yllhj.beijing.gov.cn/zwgk/ghxx/gh/201911/P020191129502960993686.pdfhttp://yllhj.beijing.gov.cn/zwgk/ghxx/gh/201911/P020191129502960993686.pdf
Chen L C, Zhu Y K, Papandreou G, Schroff F and Adam H. 2018. Encoder-decoder with atrous separable convolution for semantic image segmentation//Proceedings of the 15th European Conference on Computer Vision. Munich, Germany: Springer: 833-851[DOI: 10.1007/978-3-030-01234-2_49http://dx.doi.org/10.1007/978-3-030-01234-2_49]
Diakogiannis F I, Waldner F, Caccetta P and Wu C. 2020. ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data. ISPRS Journal of Photogrammetry and Remote Sensing, 162: 94-114[DOI:10.1016/j.isprsjprs.2020.01.013]
Helber P, Bischke B, Dengel A and Borth D. 2018. Introducing EuroSAT: a novel dataset and deep learning benchmark for land use and land cover classification//2018 IEEE International Geoscience and Remote Sensing Symposium. Valencia, Spain: IEEE: 204-207[DOI: 10.1109/IGARSS.2018.8519248http://dx.doi.org/10.1109/IGARSS.2018.8519248]
Hinton G E, Osindero S and Teh Y W. 2006. A fast learning algorithm for deep belief nets. Neural Computation, 18(7): 1527-1554[DOI:10.1162/neco.2006.18.7.1527]
Hirasawa T, Aoyama K, Tanimoto T, Ishihara S, Ozawa T, Ohnishi T, Fujishiro M, Matsuo K, Fujisaki J and Tada T. 2018. Application of artificial intelligence using a convolutional neural network for detecting gastric cancer in endoscopic images. Gastric Cancer, 21(4): 653-660[DOI:10.1007/s10120-018-0793-2]
Ioffe S and Szegedy C. 2015. Batch normalization: accelerating deep network training by reducing internal covariate shift[EB/OL].[2020-02-19].https://arxiv.org/pdf/1502.03167.pdfhttps://arxiv.org/pdf/1502.03167.pdf
Jin J, Zhu H Y, Li Z X and Sun J W. 2014. The comparison of several kinds of supervised classification methods in ENVI remote sensing image processing. Water Conservancy Science and Technology and Economy, 20(1): 146-148, 160
金杰, 朱海岩, 李子潇, 孙建伟. 2014. ENVI遥感图像处理中几种监督分类方法的比较. 水利科技与经济, 20(1): 146-148, 160[DOI:10.3969/j.issn.1006-7175.2014.01.058]
Kussul N, Lavreniuk M, Skakun S and Shelestov A. 2017. Deep learning classification of land cover and crop types using remote sensing data. IEEE Geoscience and Remote Sensing Letters, 14(5): 778-782[DOI:10.1109/LGRS.2017.2681128]
Li R R, Liu W J, Yang L, Sun S H, Hu W, Zhang F and Li W. 2018. DeepUNet: a deep fully convolutional network for pixel-level sea-land segmentation. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11(11): 3954-3962[DOI:10.1109/JSTARS.2018.2833382]
Lin T Y, Goyal P, Girshick R, He K M and Dollár P. 2017. Focal loss for dense object detection//Proceedings of 2017 IEEE International Conference on Computer Vision. Venice, Italy: IEEE: 2999-3007[DOI: 10.1109/iccv.2017.324http://dx.doi.org/10.1109/iccv.2017.324]
Liu D W, Han L and Han X Y. 2016. High spatial resolution remote sensing image classification based on deep learning. Acta Optica Sinica, 36(4): #0428001
刘大伟, 韩玲, 韩晓勇. 2016. 基于深度学习的高分辨率遥感影像分类研究. 光学学报, 36(4): #0428001[DOI:10.3788/AOS201636.0428001]
Liu H, Luo J C, Huang B, Hu X D, Sun Y W, Yang Y P, Xu N and Zhou N. 2019. DE-Net: deep encoding network for building extraction from high-resolution remote sensing imagery. Remote Sensing, 11(20): #2380[DOI:10.3390/rs11202380]
Liu P, Choo K K R, Wang L Z and Huang, F. 2017. SVM or deep learning? A comparative study on remote sensing image classification. Soft Computing, 21(23): 7053-7065[DOI:10.1007/s00500-016-2247-2]
Long J, Shelhamer E and Darrell T. 2015. Fully convolutional networks for semantic segmentation//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, USA: IEEE: 3431-3440[DOI: 10.1109/CVPR.2015.7298965http://dx.doi.org/10.1109/CVPR.2015.7298965]
Pacifici F, Chini M and Emery W J. 2009. A neural network approach using multi-scale textural metrics from very high-resolution panchromatic imagery for urban land-use classification. Remote Sensing of Environment, 113(6): 1276-1292[DOI:10.1016/j.rse.2009.02.014]
Qian Y G, Zhou W Q, Li W F and Han L J. 2015. Understanding the dynamic of greenspace in the urbanized area of Beijing based on high resolution satellite images. Urban Forestry and Urban Greening, 14(1): 39-47[DOI:10.1016/j.ufug.2014.11.006]
Ronneberger O, Fischer P and Brox T. 2015. U-Net: convolutional networks for biomedical image segmentation//Proceedings of the 18th International Conference on Medical Image Computing and Computer-assisted Intervention. Munich, Germany: Springer: 234-241[DOI: 10.1007/978-3-319-24574-4_28http://dx.doi.org/10.1007/978-3-319-24574-4_28]
Srivastava N, Hinton G, Krizhevsky A, Sutskever I and Salakhutdinov R. 2014. Dropout: a simple way to prevent neural networks from overfitting. The Journal of Machine Learning Research, 15(1): 1929-1958
Wu L J. 2006. Studies on landscape pattern and biodiversity conservation of municipal open space in Beijing. Beijing: Beijing Forestry University
吴丽娟. 2006. 北京城市绿地景观格局与生物多样性保护研究. 北京: 北京林业大学
Xu H M, Qi H, Nan K and Chen M. 2019. High-resolution remote sensing image classification by combining deep learning with nDSM. Bulletin of Surveying and Mapping, (8): 63-67
许慧敏, 齐华, 南轲, 陈敏. 2019. 结合nDSM的高分辨率遥感影像深度学习分类方法. 测绘通报, (8): 63-67[DOI:10.13474/j.cnki.11-2246.2019.0253]
Zhang Z X, Liu Q J and Wang Y H. 2018. Road extraction by deep residual U-Net. IEEE Geoscience and Remote Sensing Letters,15(5): 749-753[DOI:10.1109/LGRS.2018.2802944]
Zheng M G, Cai Q G, Qin M Z and Yue T X. 2006. A new approach to accuracy assessment of classifications of remotely sensed data. Journal of Remote Sensing, 10(1): 39-48
郑明国, 蔡强国, 秦明周, 岳天祥. 2006.一种遥感影像分类精度检验的新方法. 遥感学报, 10(1): 39-48[DOI:10.11834/jrs.20060107]
相关作者
相关机构
京公网安备11010802024621