一种融合空洞卷积与池化模型的遥感影像水体提取方法

doi:10.16180/j.cnki.issn1007-7820.2025.03.006

电子科技 ›› 2025, Vol. 38 ›› Issue (3): 40-46.doi: 10.16180/j.cnki.issn1007-7820.2025.03.006

一种融合空洞卷积与池化模型的遥感影像水体提取方法

赵云飞¹, 薛存金²()

1.昆明理工大学国土资源工程学院,云南昆明 650093
2.中国科学院空天信息创新研究院数字地球重点实验室,北京 100094

收稿日期:2023-09-05 修回日期:2023-09-24 出版日期:2025-03-15 发布日期:2025-03-11
通讯作者: 薛存金(1979-),男,E-mail:2862433266@qq.com,博士,研究员。研究方向:人工智能应用、时空数据挖掘。
作者简介:赵云飞(1996-),男,硕士研究生。研究方向:人工智能应用。
基金资助:
国家自然科学基金(42376193)

A Remote Sensing Image Water Extraction Method by Combining Atrous Convolution and Pooling Models

ZHAO Yunfei¹, XUE Cunjin²()

1. Faculty of Land Resource Engineering,Kunming University of Science and Technology,Kunming 650093,China
2. Key Laboratory of Digital Earth Science,Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China

Received:2023-09-05 Revised:2023-09-24 Online:2025-03-15 Published:2025-03-11
Supported by:
National Natural Science Foundation of China(42376193)

摘要/Abstract

摘要：

植被、阴影和云层等同谱异物物体的干扰导致遥感影像水体提取完整性较低、提取效果差。文中提出一种融合多层次空洞卷积和池化模型的遥感影像水体提取模型MAP_UNet(A UNet of Combining Multi Atrous Convolution and Pooling Model)。该模型以UNet为基准编解码网络,提取水体的不同尺寸特征,引入双递归残差模块防止出现深层网络梯度消失现象,并使用多模块来融合空间空洞卷积和最大池化以捕捉更大范围的特征信息,进一步加强相邻尺度的特征语义关系。为验证所提方法的有效性与先进性,利用高分辨率可见光遥感影像数据进行实验,并与公开深度学习语义分割算法进行对比。实验结果表明,MAP_UNet模型在提取精度和防止同谱异物体误检方面都取得了较好效果,其精确率、召回率、F1-Score和MIoU(Mean Intersection over Union)分别达96.20%、92.64%、87.27%和89.10%,相比UNet(U-shaped Network)、UNet++和UNet_ASPP(UNet with Atrous Spatial Pyramid Pooling Network)网络均有较大提升。

关键词: 水体提取, 深度学习, 神经网络, 空洞卷积, 语义分割, UNet, 遥感影像

Abstract:

The interference of spectral foreign objects such as vegetation, shadows and clouds leads to the low integrity and poor extraction effect of remote sensing images. A remote sensing image water extraction model MAP_UNet(A UNet of Combining Multi Atrous Convolution and Pooling Model) is proposed in this study, which combines multi-level cavity convolution and pooling model. The model uses UNet as the reference codec network to extract different dimensional features of water bodies, introduces double recursive residual module to prevent gradient disappearance of deep network, and uses multiple modules to integrate spatial cavity convolution and maximum pooling to capture a larger range of feature information and further strengthen the feature semantic relationship of adjacent scales. In order to verify the effectiveness and advance of the proposed method, experiments are carried out using high-resolution visible remote sensing image data and the results are compared with open deep learning semantic segmentation algorithm. The experimental results show that the MAP_UNet model has achieved good results in extracting accuracy and preventing misdetection of different objects of the same spectrum. The accuracy rate, recall rate, F1-Score and MIoU(Mean Intersection over Union) are 96.20%, 92.64%, 87.27% and 89.10%, respectively. Compared with UNet(U-shaped Network), UNet++ and UNet ASPP(UNet with Atrous Spatial Pyramid Pooling Network) networks, the performance of proposed method has significant improvement.

Key words: water extraction, deep learning, neural network, atrous convolution, semantic segmentation, UNet, remote sensing image

中图分类号:

TP391

赵云飞, 薛存金. 一种融合空洞卷积与池化模型的遥感影像水体提取方法[J]. 电子科技, 2025, 38(3): 40-46.

ZHAO Yunfei, XUE Cunjin. A Remote Sensing Image Water Extraction Method by Combining Atrous Convolution and Pooling Models[J]. Electronic Science and Technology, 2025, 38(3): 40-46.

图/表 8

图1

图2

图3

图4

表1

图5

图6

表2

参考文献 20

[1]	苏龙飞, 李振轩, 高飞, 等. 遥感影像水体提取研究综述[J]. 国土资源遥感, 2021, 33(1):9-19.
	Su Longfei, Li Zhenxuan, Gao Fei, et al. A review of remote sensing image water extraction[J]. Remote Sensing for Natural Resources, 2021, 33(1):9-19.
[2]	周鹏, 谢元礼, 蒋广鑫, 等. 遥感影像水体信息提取研究进展[J]. 遥感信息, 2020, 35(5):9-18.
	Zhou Peng, Xie Yuanli, Jiang Guangxin, et al. Advances on water body information extraction from remote sensing imagery[J]. Remote Sensing Information, 2020, 35(5):9-18.
[3]	刘璐, 张洪艳, 张良培. 基于光谱加权低秩矩阵分解的高光谱影像去噪方法[J]. 电子科技, 2020, 33(05):21-27.
	Liu Lu, Zhang Hongyan, Zhang Liangpei. Hyperspectral image denoising via spectral weighted low-rank matrix approximate[J]. Electronic Science and Technology, 2020, 33(5):21-27.
[4]	温泉, 李璐, 熊立, 等. 基于深度学习的遥感图像水体提取综述[J]. 自然资源遥感, 2024, 36(3):57-71.
	Wen Quan, Li Lu, Xiong Li, et al. A review of water extraction from remote sensing images based on deep learning[J]. Remote Sensing for Natural Resources, 2024, 36(3):57-71.
[5]	Cao H, Zhang H, Wang C, et al. Operational flood detection using sentinel-1 SAR data over large areas[J]. Water, 2019, 11(4):786-806.
[6]	Yao F F, Wang J D, Wang C, et al. Constructing long-term high-frequency time series of global lake and reservoir areas using Landsat imagery[J]. Remote Sensing of Environment, 2019, 232(10):1113-1133.
[7]	Ahmad K S, Hossain F, Eldardiry H, et al. A fusion approach for water area classification using visible, near infrared and synthetic aperture radar for south Asian conditions[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(4):2471-2480.
[8]	吴梦娟. 基于深度学习的高分卫星数据水体提取研究[D]. 南京: 南京信息工程大学,2020:1-74.
	Wu Mengjuan. Research onwater extraction of GF-1 satellite data based on deep learning[D]. Nanjing: Nanjing University of Information Science and Technology,2020:1-74.
[9]	吴瑞娟, 何秀凤. 高分雷达与光学影像融合的滨海湿地变化检测[J]. 测绘科学, 2020, 45(11):93-100.
	Wu Ruijuan, He Xiufeng. Coastal wetland change detection using fusion of high of resolution radar and optical images[J]. Science of Surveying and Mapping, 2020, 45(11): 93-100.
[10]	洪亮, 黄雅君, 杨昆, 等. 复杂环境下高分二号遥感影像的城市地表水体提取[J]. 遥感学报, 2019, 23(5):871-882.
	Hong Liang, Huang Yajun, Yang Kun, et al. Study on urban surface water extraction from heterogeneous environments using GF-2 remotely sensed images[J]. National Remote Sensing Bulletin, 2019, 23(5):871-882.
[11]	汪权方, 张梦茹, 张雨, 等. 基于视觉注意机制的大范围水体信息遥感智能提取[J]. 计算机应用, 2020, 40(4):1038-1044.
	Wang Quanfang, Zhang Mengru, Zhang Yu, et al. Intelligent extraction of remote sensing information on large-scale water based on visual attention mechanism[J]. Journal of Computer Applications, 2020, 40(4):1038-1044.
[12]	周婷, 汪炎, 邹俊, 等. 基于PCA和SVM的遥感影像水体提取方法及验证[J]. 水资源保护, 2023, 39(2):180-189.
	Zhou Ting, Wang Yan, Zou Jun, et al. PCA and SVM-based algorithm of water area extraction from remote sensing images and its verification[J]. Water Resources Protection, 2023, 39(2):180-189.
[13]	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.
	Chen Qian, Zheng Lijuan, Li Xiaojuan, 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.
[14]	梁泽毓, 吴艳兰, 杨辉, 等. 基于密集连接全卷积神经网络的遥感影像水体全自动提取方法[J]. 遥感信息, 2020, 35(4):68-77.
	Liang Zeyu, Wu Yanlan, Yang Hui, et al. Fully-automatic water extraction method for remote sensing imagery based on densely connected fully convolutional neural network[J]. Remote Sensing Information, 2020, 35(4):68-77.
[15]	郑泰皓, 王庆涛, 李家国, 等. 基于深度学习的高分六号影像水体自动提取[J]. 科学技术与工程, 2021, 21(4):1459-1470.
	Zheng Taihao, Wang Qingtao, Li Jiaguo, et al. Automatic water extraction from GF-6 image based on deep larning[J]. Science Technology and Engineering, 2021, 21(4):1459-1470.
[16]	李鑫伟, 李彦胜, 张永军. 弱监督深度语义分割网络的多源遥感影像水体检测[J]. 中国图象图形学报, 2021, 26(12):3015-3026.
	Li Xinwei, Li Yansheng, Zhang Yongjun. Weakly supervised deep semantic segmentation network for water body extraction based on multi-source remote sensing imagery[J]. Journal of Image and Graphics, 2021, 26(12):3015-3026.
[17]	Ronneberger O, Fischer P, Brox T. U-Net: Convolutional networks for biomedical image segmentation[C]. Munich:In Medical Image Computing and Computer Assisted Intervention:The Eighteenth International Conference,2015:234-241.
[18]	Zhou Z W, Rahman Siddiquee M M, Tajbakhsh N, et al. UN-et++:A nested UNET architecture for medical image segmentation[J]. IEEE Transactions on Medical Imaging, 2018, 39(9):1856-1867.
[19]	Badrinarayanan V, Kendall A, Cipolla R. SegNet:A deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12):2481-2495.
[20]	Yu F, Koltun V, Funkhouser T. Dilated residual networks[C]. Honolulu: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2017:472-480.

模型	Precision	Recall	F1-Score	MIoU
UNet	0.953 531	0.914 480	0.862 469	0.869
UNet++	0.953 198	0.891 954	0.858 620	0.850
UNet_ASPP	0.949 169	0.907 647	0.850 527	0.878
MAP_UNet	0.961 960	0.926 421	0.872 708	0.891

一种融合空洞卷积与池化模型的遥感影像水体提取方法

A Remote Sensing Image Water Extraction Method by Combining Atrous Convolution and Pooling Models

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 20

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	袁德品, 赵亮, 葛宪生. 基于复数深度神经网络的电磁信号调制识别[J]. 电子科技, 2025, 38(3): 1-6.
[2]	王紫祎, 陈世平. 基于图对比学习的自监督网络流量检测模型[J]. 电子科技, 2025, 38(3): 22-31.
[3]	陈宇洋, 李峰. 融合CNN与Transformer的视网膜OCT图像积液分割方法[J]. 电子科技, 2025, 38(3): 47-59.
[4]	郑方亮, 王延年, 廉继红, 阮佩. 基于条件先验Swin Transformer的人脸图像超分辨重建[J]. 电子科技, 2025, 38(2): 35-41.
[5]	叶翔宇, 林晓会, 丁江乔, 解维坤. 基于优化BPNN的FPGA内嵌高速接口总抖动预测方法[J]. 电子科技, 2025, 38(2): 70-77.
[6]	赖颖, 巨志勇, 叶雨新. 基于改进YOLOv4的车辆检测算法[J]. 电子科技, 2025, 38(1): 81-87.
[7]	何志强, 孙占全. 基于Swin-Transformer的颈动脉超声图像斑块分割[J]. 电子科技, 2024, 37(9): 48-56.
[8]	杨潇, 李高磊. 半监督图节点分类任务的清洁标签后门植入[J]. 电子科技, 2024, 37(9): 57-63.
[9]	卢臻阳. 网络安全态势预测技术研究综述[J]. 电子科技, 2024, 37(8): 92-96.
[10]	韩瀛昊, 李菲菲. 基于关键区域特征提取与二阶段分类网络的场景识别方法[J]. 电子科技, 2024, 37(7): 25-32.
[11]	何幸, 黄永明, 朱勇. 基于改进YOLOv5的路面坑洼检测方法[J]. 电子科技, 2024, 37(7): 53-59.
[12]	赵旭, 胡德敏. 多路并行多尺度特征复用的遥感图像超分辨率[J]. 电子科技, 2024, 37(6): 61-68.
[13]	叶雨新, 巨志勇, 赖颖. 融合感受野增强和注意力机制的交通标志检测算法[J]. 电子科技, 2024, 37(6): 8-16.
[14]	袁傲, 齐金鹏, 贾灿, 薛宇鑫, 郭阳阳. 一种基于深度学习的时序病变数据段分类方法[J]. 电子科技, 2024, 37(6): 84-91.
[15]	朱子豪, 宋燕. 融合注意力和胶囊池化的轻量型胶囊网络[J]. 电子科技, 2024, 37(5): 1-8.