基于HRNet的高分辨率遥感影像道路提取方法

doi:10.12305/j.issn.1001-506X.2024.04.04

系统工程与电子技术 ›› 2024, Vol. 46 ›› Issue (4): 1167-1173.doi: 10.12305/j.issn.1001-506X.2024.04.04

基于HRNet的高分辨率遥感影像道路提取方法

陈雪梅¹^,², 刘志恒¹^,²^,*, 周绥平¹, 余航¹, 刘彦明¹

1. 西安电子科技大学空间科学与技术学院, 陕西西安 710126
2. 自然资源部矿山地质灾害成灾机理与防控重点实验室, 陕西西安 710054

收稿日期:2023-01-19 出版日期:2024-03-25 发布日期:2024-03-25
通讯作者: 刘志恒
作者简介:陈雪梅 (1999—), 女, 硕士研究生, 主要研究方向为遥感图像处理
刘志恒 (1990—), 男, 讲师, 博士, 主要研究方向为遥感科学与技术
周绥平 (1968—), 男, 教授, 博士, 主要研究方向为分布式虚拟环境、人工智能与空间科学大数据分析
余航 (1982—), 男, 副教授, 博士, 主要研究方向为图像分割与计算机视觉
刘彦明 (1966—), 男, 教授, 博士, 主要研究方向为电子通信与雷达探测
基金资助:
自然资源部矿山地质灾害成灾机理与防控重点实验室项目(2022-08);陕西省自然科学基础研究计划(2023-JC-QN-0299);中央高校基本科研业务费(XJS221307)

Road extraction from high-resolution remote sensing images based on HRNet

Xuemei CHEN¹^,², Zhiheng LIU¹^,²^,*, Suiping ZHOU¹, Hang YU¹, Yanming LIU¹

1. School of Aerospace Science & Technology, Xidian University, Xi'an 710126, China
2. Key Laboratory of Mine Geological Hazards Mechanism and Control, Ministry of Natural Resources, Xi'an 710054, China

Received:2023-01-19 Online:2024-03-25 Published:2024-03-25
Contact: Zhiheng LIU

摘要/Abstract

摘要：

高分辨率遥感影像中，传统的道路提取方法存在着精度低、鲁棒性低的问题，提出基于高分辨率网络(high-resolution net, HRNet)实现高分辨率遥感影像道路分割。对HRNet进行改进，将相同分辨率的HRNet子网的输出与输出层结果进行拼接并输入非局部块，两个损失函数Cross-entropy Loss和Dice Loss用来解决道路数据集样本不平衡问题。实验结果表明，改进的HRNet在公开的CHN6-CUG道路数据集上的分割性能与其他方法相比对道路的提取效果更好，在召回率、均交并比和F1分数3个方面分别达到了97.65%、84.91%和97.25%。

关键词: 高分辨率网络, 非局部块, 遥感影像, 深度学习

Abstract:

In response to the problems of low accuracy and weak robustness in traditional road extraction from high-resolution remote sensing images, a high-resolution net (HRNet) based approach is proposed to achieve road segmentation in high-resolution remote sensing images. This article improves HRNet by concatenating the output of HRNet subnets of the same resolution with the output layer results and inputting the above into non-local block. The two loss functions, Cross-entropy Loss and Dice Loss, are used to solve the problem of imbalanced road dataset samples. The experimental results show that the improved HRNet performs better in road extraction on the publicly available CHN6-CUG road dataset compared to other methods, achieving 97.65%, 84.91%, and 97.25% respectively in recall, mean intersection over union (MIoU), and F1 score.

Key words: high-resolution net (HRNet), non-local block, remote sensing image, deep learning

中图分类号:

TP751

陈雪梅, 刘志恒, 周绥平, 余航, 刘彦明. 基于HRNet的高分辨率遥感影像道路提取方法[J]. 系统工程与电子技术, 2024, 46(4): 1167-1173.

Xuemei CHEN, Zhiheng LIU, Suiping ZHOU, Hang YU, Yanming LIU. Road extraction from high-resolution remote sensing images based on HRNet[J]. Systems Engineering and Electronics, 2024, 46(4): 1167-1173.

图/表 8

图1

图2

图3

表1

表2

图4

图5

表3

参考文献 38

1	刘洋, 康健, 管海燕, 等. 基于双注意力残差网络的高分遥感影像道路提取模型[J]. 地球信息科学学报, 2023, 25 (2): 396- 408.
	LIU Y , KANG J , GUAN H Y , et al. Road extraction model of high-resolution remote sensing images based on dual-attention residual network[J]. Journal of Geo-Information Science, 2023, 25 (2): 396- 408.
2	梁茜亚, 王卷乐, 李朋飞, 等. 基于高分一号(GF-1)影像的蒙古高原干旱半干旱地区自然道路提取研究——以蒙古国古尔班特斯苏木为例[J]. 自然资源遥感, 2022, 35 (2): 122- 131.
	LIANG Q Y , WANG J L , LI P F , et al. Research on natural roads extraction in arid and semiarid regions of the Mongolian Plateau based on GF-1 images-take the Gurvantes Soum, Mongolia as an example[J]. Remote Sensing for Natural Resources, 2022, 35 (2): 122- 131.
3	DAI L , ZHANG G Y , ZHANG R T . RADANet: road augmented deformable attention network for road extraction from complex high-resolution remote-sensing images[J]. IEEE Trans.on Geoscience and Remote Sensing, 2023, 61, 5602213.
4	ZHANG X , ZHANG C K , LI H M , et al. A road extraction method based on high resolution remote sensing image[J]. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2020, 671- 676.
5	LIU L Y , DONG Y Y , HUANG W J , et al. Enhanced regional monitoring of wheat powdery mildew based on an instance-based transfer learning method[J]. Remote Sensing, 2019, 11 (3): 298. doi: 10.3390/rs11030298
6	MEGHANA I, MEGHANA J D N V L, JAYARAMAN R. Smart attendance management system using radio frequency identification[C]//Proc. of the International Conference on Communication and Signal Processing, 2020: 1045-1049.
7	戴激光, 王杨, 杜阳, 等. 光学遥感影像道路提取的方法综述[J]. 遥感学报, 2020, 24 (7): 804- 823.
	DAI J G , WANG Y , DU Y , et al. Development and prospect of road extraction method for optical remote sensing image[J]. National Remote Sensing Bulletin, 2020, 24 (7): 804- 823.
8	DAI J G , MA R C , GONG L T , et al. A model-driven-to-sample-driven method for rural road extraction[J]. Remote Sensing, 2021, 13 (8): 1417. doi: 10.3390/rs13081417
9	LIN X G , ZHANG R , SHEN J . A template-matching based approach for extraction of roads from very high-resolution remotely sensed imagery[J]. International Journal of Image and Data Fusion, 2012, 3 (2): 149- 168. doi: 10.1080/19479832.2011.642413
10	TAN H , SHEN Z M , DAI J G . Semi-automatic extraction of rural roads under the constraint of combined geometric and texture features[J]. ISPRS International Journal of Geo-Information, 2021, 10 (11): 754. doi: 10.3390/ijgi10110754
11	YANG K L , CUI W H , SHI S , et al. Semi-automatic method of extracting road networks from high-resolution remote-sensing images[J]. Applied Sciences, 2022, 12 (9): 4705. doi: 10.3390/app12094705
12	GUO Q , WANG Z P . A self-supervised learning framework for road centerline extraction from high-resolution remote sensing images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13, 4451- 4461. doi: 10.1109/JSTARS.2020.3014242
13	LI M M , STEIN A , BIJKER W , et al. Region-based urban road extraction from VHR satellite images using binary partition tree[J]. International Journal of Applied Earth Observation and Geo-information, 2016, 44 (9): 217- 225.
14	LIN X G , XIE W H , ZHANG L B , et al. Semi-automatic road extraction from high resolution satellite images by template matching using Kullback-Leibler divergence as a similarity measure[J]. International Journal of Image and Data Fusion, 2022, 13 (4): 316- 336. doi: 10.1080/19479832.2022.2121767
15	房玉品, 王小鹏, 李新娜. 基于自适应形态学的遥感图像道路提取[J]. 激光与光电子学进展, 2022, 59 (16): 135- 142.
	FANG Y P , WANG X P , LI X N . Road extraction from remote sensing images based on adaptive morphology[J]. Laser & Optoelectronics Progress, 2022, 59 (16): 135- 142.
16	林鹏, 阮仁宗, 王玉强, 等. 一种基于面向对象的城镇道路自动提取方法研究[J]. 地理与地理信息科学, 2016, 32 (1): 42.
	LIN P , RUAN R Z , WANG Y Q , et al. Research on extraction of road based on object oriented in an urban context[J]. Geography and Geo-Information Science, 2016, 32 (1): 42.
17	ABDOLLAHI A , PRADHAN B , ALAMRI A . VNet: an end-to-end fully convolutional neural network for road extraction from high-resolution remote sensing data[J]. IEEE Access, 2020, 8, 179424- 179436. doi: 10.1109/ACCESS.2020.3026658
18	WANG Y S , SEO J H , JEON T Y . NL-LinkNet: toward lighter but more accurate road extraction with nonlocal operations[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19, 3000105.
19	DING L , BRUZZONE L . DiResNet: direction-aware residual network for road extraction in VHR remote sensing images[J]. IEEE Trans.on Geoscience and Remote Sensing, 2021, 59 (12): 10243- 10254. doi: 10.1109/TGRS.2020.3034011
20	RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation[C]//Proc. of the International Conference on Medical Image Computing and Computer-Assisted Intervention, 2015: 234-241.
21	ZHOU Z W , RAHMAN S M M , TAJBAKHSH N , et al. UNet++: redesigning skip connections to exploit multiscale features in image segmentation[J]. IEEE Trans.on Medical Imaging, 2019, 39 (6): 1856- 1867.
22	ZHAO H S, SHI J P, QI X J, et al. Pyramid scene parsing network[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 6230-6239.
23	CHEN L C, ZHU Y, PAPANDREOU G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Proc. of the Europeon Conference on Computer Vision, 2018: 833-851.
24	赵凌虎, 袁希平, 甘淑, 等. 改进Deeplabv3+的高分辨率遥感影像道路提取模型[J]. 自然资源遥感, 2023, 35 (1): 107- 114.
	ZHAO L H , YUAN X P , GAN S , et al. An information extraction model of roads from high-resolution remote sensing images based on improved Deeplabv3+[J]. Remote Sensing for Natural Resources, 2023, 35 (1): 107- 114.
25	SANDLER M, HOWARD A, ZHU M, et al. MobileNetV2: inverted residuals and linear bottlenecks[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 4510-4520.
26	龙伊娜, 谷玉海, 吴文昊, 等. 基于改进D-Linknet的高分遥感影像道路提取方法[J]. 激光杂志, 2023, 44 (5): 162- 168.
	LONG Y N , GU Y H , WU W H , et al. Road extraction method of high resolution remote sensing image based on improved D-Linknet[J]. Laser Journal, 2023, 44 (5): 162- 168.
27	GIUSTI A, CIREȘAN D, MASCI J, et al. Fast image scanning with deep max-pooling convolutional neural networks[C]//Proc. of the IEEE International Conference on Image Processing, 2013.
28	WANG J D , KE S , CHENG T H , et al. Deep high-resolution representation learning for visual recognition[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 2020, 33 (10): 3349- 3364.
29	WANG X, GIRSHICK R, GUPTA A, et al. Non-local neural networks[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 7794-7803.
30	SUN K, XIAO B, LIU D, et al. Deep high-resolution representation learning for human pose estimation[C]//Proc. of the 32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 5686-5696.
31	WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proc. of the European Conference on Computer Vision, 2018: 3-19.
32	JADON S. A survey of loss functions for semantic segmentation[C]//Proc. of the IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, 2020.
33	MA Y D, LIU Q, QIAN Z B. Automated image segmentation using improved PCNN model based on cross-entropy[C]//Proc. of the International Symposium on Intelligent Multimedia, Video and Speech Processing, 2004: 743-746.
34	SUDRE C H, LI W, VERCAUTEREN T, et al. Generalised dice overlap as a deep learning loss function for highly unba-lanced segmentations[C]//Proc. of the International Workshop on Deep Learning in Medical Image Analysis, 2017: 240-248.
35	ZHU Q Q , ZHANG Y N , WANG L Z , et al. A global context-aware and batch-independent network for road extraction from VHR satellite imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2021, 175 (12): 353- 365.
36	SHELHAMER E , LONG J , DARRELL T . Fully convolutional networks for semantic segmentation[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 2016, 39 (4): 640- 651.
37	ZHOU L C, ZHANG C, WU M. D-LinkNet: linkNet with pretrained encoder and dilated convolution for high resolution satellite imagery road extraction[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2018: 192-1924.
38	TIELEMAN T , HINTON G . Lecture 6.5-rmsprop: divide the gradient by a running average of its recent magnitude[J]. COURSERA: Neural Networks for Machine Learning, 2012, 4 (2): 26- 31.

网络模型	Recall	Precision	MIoU	F1
HRNet	94.80	98.19	79.28	96.46
本文方法-N^*	96.24	95.56	83.25	96.81
本文方法	97.26	96.85	84.91	97.25

网络模型	Recall	Precision	MIoU	F1
U-Net	95.55	97.42	80.72	96.47
FCN	93.37	98.09	73.97	95.66
PSPNet	93.56	98.59	75.59	96.01
DeeplabV3+	95.08	98.21	79.96	96.62
D-LinkNet50	89.71	87.45	54.16	87.91
HRNet	94.80	98.19	79.28	96.46
本文方法	97.26	96.85	84.91	97.25

网络模型	参数量/MB	计算量	预测时间/s
U-Net	13.40	124.49	0.49
FCN	23.64	28.99	0.20
PSPNet	46.71	184.73	0.59
DeeplabV3+	40.47	76.54	0.55
D-LinkNet50	217.65	120.39	0.78
HRNet	9.64	18.66	0.24
本文方法	12.38	60.46	0.46

[1]	张天文, 张晓玲, 邵子康, 曾天娇. 基于掩模注意型交互的SAR舰船实例分割[J]. 系统工程与电子技术, 2024, 46(3): 831-838.
[2]	施端阳, 林强, 胡冰, 杜小帅. 基于YOLO的航管一次雷达目标检测方法[J]. 系统工程与电子技术, 2024, 46(1): 143-151.
[3]	汪萌, 诸兵. 不确定性建模在2D和3D目标检测中的应用[J]. 系统工程与电子技术, 2023, 45(8): 2370-2376.
[4]	邵凯, 杜自群, 王光宇. 基于模型剪枝动态调整压缩率的CSI反馈方法[J]. 系统工程与电子技术, 2023, 45(8): 2615-2622.
[5]	崔天舒, 王栋, 黄振. 面向空间认知通信的轻量化网络自动调制分类方法[J]. 系统工程与电子技术, 2023, 45(7): 2220-2226.
[6]	姜雨, 袁琪, 胡志韬, 吴薇薇, 顾欣. 基于气象因素的机场进离港延误预测[J]. 系统工程与电子技术, 2023, 45(6): 1722-1731.
[7]	陈洋, 廖灿辉, 张锟, 刘建, 王鹏举. 基于自监督对比学习的信号调制识别算法[J]. 系统工程与电子技术, 2023, 45(4): 1200-1206.
[8]	张晔, 侯毅, 欧阳克威, 周石琳. 单变量序列数据分类方法综述[J]. 系统工程与电子技术, 2023, 45(2): 313-335.
[9]	邵正途, 许登荣, 徐文利, 王晗中. 基于LSTM和残差网络的雷达有源干扰识别[J]. 系统工程与电子技术, 2023, 45(2): 416-423.
[10]	陈任飞, 彭勇, 李忠文. 基于持续无监督域适应策略的水面漂浮物目标检测方法[J]. 系统工程与电子技术, 2023, 45(11): 3391-3401.
[11]	陶臣嵩, 陈思伟, 肖顺平. 基于深度学习模型的SAR图像间歇采样转发干扰检测[J]. 系统工程与电子技术, 2023, 45(11): 3465-3473.
[12]	何诗扬, 汪玲, 朱岱寅, 钱君. 基于时空记忆解耦RNN的雷暴预测方法[J]. 系统工程与电子技术, 2023, 45(11): 3474-3480.
[13]	李浩然, 熊伟, 崔亚奇. 基于深度特征融合的SAR图像与AIS信息关联方法[J]. 系统工程与电子技术, 2023, 45(11): 3491-3497.
[14]	刘子超, 王江, 何绍溟. 基于深度学习的时间角度控制制导律[J]. 系统工程与电子技术, 2023, 45(11): 3579-3587.
[15]	季策, 宋博翰, 耿蓉, 梁敏骏. 快时变信道下基于深度学习的OFDM系统信道估计[J]. 系统工程与电子技术, 2023, 45(11): 3649-3655.

基于HRNet的高分辨率遥感影像道路提取方法

Road extraction from high-resolution remote sensing images based on HRNet

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 38

相关文章 15

编辑推荐

Metrics

本文评价