Greenhouse segmentation of remote sensing images based on deep learning
-
摘要:
大棚面积统计是蔬菜业进行科学管理的客观需求，通过深度学习自动获取大棚面积能更好地降低人工成本.为了准确获取大棚面积，本文使用基于深度学习的模型框架来解决这一问题.首先，使用金字塔池化模块和链式池化模块分别处理大尺度和小尺度的高层特征，在保持分辨率和避免退化的情况下提取多尺度的语义信息；然后，使用递归残差模块交替融合高层和低层信息，进一步优化分割结果。通过和三种较新的算法进行比较，实验结果表明本文的方法取得了最优的结果.
Abstract:Greenhouse area statistics is the objective requirement of scientific management of vegetable industry. It can reduce labor costs better to acquire greenhouse area automatically through deep learning. In order to obtain the greenhouse area accurately, a framework based on deep learning is proposed to solve this problem. First of all, pyramid pooling module and chain pooling module are used to deal with high-level features of large scale and small scale respectively, and multi-scale semantic information is extracted under the condition of maintaining resolution and avoiding degradation. Then, the recursive residual module is used to alternately fuse the high-level and low-level1 information to further optimize the segmentation results. Comparing with three newer algorithms, the experimental results show that the proposed method has achieved the best results..
-
Key words:
- remote sensing /
- greenhouse segmentation /
- deeplearning /
- multi-scale /
- recursive residual learning
-
表 1 实验结果对比
mIoU MAE Precision Recall F-Measure DeepLabv3 0.631 9 0.153 1 0.773 5 0.7262 0.762 0 UNet 0.783 2 0.091 0 0.800 8 0.872 2 0.816 2 PSPNet 0.796 9 0.056 6 0.840 6 0.818 3 0.835 3 OUR 0.807 2 0.051 7 0.844 5 0.837 1 0.842 8 
下载: 导出CSV
-
[1] 李静, 赵庚星, 李涛, 等. TM影像中大棚菜地信息提取技术研究[J].水土保持学报, 2004, 18(1): 126-129. DOI: 10.3321/j.issn:1009-2242.2004.01.033.LI J, ZHAO G X, LI T, et al. Study on technique of extracting greenhouse vegetable information from Landsat TM image[J]. Journal of Soil and Water Conservation, 2004, 18(1): 126-129. DOI: 10.3321/j.issn:1009-2242.2004.01.033. [2] 黄振国, 陈仲新, 刘芳清, 等.基于HJ-1影像的大棚菜地遥感监测技术研究--以山东寿光市为例[J].中国农业资源与区划, 2013, 34(5): 102-106. DOI: 10.7621/cjarrp.1005-9121.20130516.HUANG Z G, CHEN Z X, LIU F Q, et al. Monitoring of greenhouse vegetables land using HJ-1 remotely-sensed imagery[J]. Chinese Journal of Agricultural Resources and Regional Planning, 2013, 34(5): 102-106. DOI: 10.7621/cjarrp.1005-9121.20130516. [3] 徐旭.基于图像特征的遥感影像水体自动提取方案研究[J].安徽建筑, 2019, 26(9): 50-51. DOI: 10.16330/j.cnki.1007-7359.2019.09.020. [4] 冯朝晖, 李芹, 韩留生, 等.改进分水岭方法下的遥感图像水体提取[J].测绘通报, 2019(6): 11-15. DOI: 10.13474/j.cnki.11-2246.2019.0175.FENG Z H, LI Q, HAN L S, et al. Remote sensing image water body extraction based on improved watershed[J]. Bulletin of Surveying and Mapping, 2019(6): 11-15. DOI: 10.13474/j.cnki.11-2246.2019.0175. [5] 陈前, 郑利娟, 李小娟, 等.基于深度学习的高分遥感影像水体提取模型研究[J].地理与地理信息科学, 2019, 35(4): 43-49. DOI: 10.3969/j.issn.1672-0504.2019.04.007.CHEN Q, ZHENG L J, LI X J, et al. Water body extraction from high-resolution satellite remote sensing images based on deep learning[J]. Geography and Geo-Information Science, 2019, 35(4): 43-49. DOI: 10.3969/j.issn.1672-0504.2019.04.007. [6] 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, MA, USA: IEEE, 2015: 3431-3440. DOI: 10.1109/CVPR.2015.7298965. [7] ZHENG S, JAYASUMANA S, ROMERA-PAREDES B, et al. Conditional random fields as recurrent neural networks[C]//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago, Chile: IEEE, 2015: 1529-1537. DOI: 10.1109/ICCV.2015.179. [8] SUN K, ZHAO Y, JIANG B R, et al. High-resolution representations for labeling pixels and regions[J]. arXiv: 1904.04514, 2019. https://www.researchgate.net/publication/332316206_High-Resolution_Representations_for_Labeling_Pixels_and_Regions [9] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(4): 834-848. DOI: 10.1109/TPAMI.2017.2699184. [10] LIN G S, MILAN A, SHEN C H, et al. RefineNet: multi-path refinement networks for high-resolution semantic segmentation[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, HI, USA: IEEE, 2017: 5168-5177. DOI: 10.1109/CVPR.2017.549. [11] CHEN L C, PAPANDREOU G, SCHROFF F, et al. Rethinking atrous convolution for semantic image segmentation[EB/OL].[2019-05-23]. https://arxiv.org/pdf/1706.05587.pdf. [12] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE, 2016: 770-778. DOI: 10.1109/CVPR.2016.90. [13] ZHOU B L, KHOSLA A, LAPEDRIZA A, et al. Object detectors emerge in deep scene CNNs[J]. arXiv: 1412.6856, 2014. https://arxiv.org/abs/1412.6856 [14] RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation[C]//Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention. Munich, Germany: Springer, 2015: 234-241. DOI: 10.1007/978-3-319-24574-4_28. [15] DENG Z J, HU X W, ZHU L, et al. R3Net: recurrent residual refinement network for saliency detection[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence. California: International Joint Conferences on Artificial Intelligence Organization, 2018: 684-690. DOI: 10.24963/ijcai.2018/95. [16] CHEN L C, ZHU Y K, PAPANDREOU G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[EB/OL].[2019-05-28]. https://github.com/tensorflow/models/tree/master/. [17] ZHAO H S, SHI J P, QI X J, et al. Pyramid scene parsing network[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, HI, USA: IEEE, 2017: 2881-2890. DOI: 10.1109/CVPR.2017.660. -
下载:
点击查看大图
图(4) / 表(1)
计量
- 文章访问数: 416
- HTML全文浏览量: 211
- PDF下载量: 45
- 被引次数: 0
