PDF Download

[1]ZHAO Zilin,HAN Lei,CHEN Rui,et al.Positive and Negative Terrain Segmentation in the Loess Hilly and Gully Region Based on Deep Learning[J].Research of Soil and Water Conservation,2023,30(05):21-30.[doi:10.13869/j.cnki.rswc.2023.05.015.]

Copy

Positive and Negative Terrain Segmentation in the Loess Hilly and Gully Region Based on Deep Learning

Research of Soil and Water Conservation[ISSN 1005-3409/CN 61-1272/P] Volume: 30 Number of periods: 2023 05 Page number: 21-30 Column: Public date: 2023-08-10

Title:: Positive and Negative Terrain Segmentation in the Loess Hilly and Gully Region Based on Deep Learning

Author(s):: ZHAO Zilin¹, HAN Lei^2,3, CHEN Rui¹, ZHAO Yonghua², LI Yabei¹, KANG Hongliang²; (1.School of Earth Sciences and Resources, Chang'an University, Xi'an 710054, China; 2.School of Land Engineering, Shaanxi Key Laboratory of Land consolidation, Chang'an University, Xi'an 710054, China; 3.State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences, Xi'an 710061, China)

Keywords:: positive and negative terrain segmentation; deep learning; Unet; residual block(RB); convolutional block attention module(CBAM)

CLC:: P208

DOI:: 10.13869/j.cnki.rswc.2023.05.015.

Abstract:: [Objective] Based on deep learning, a large-scale and high-precision positive and negative terrains segmentation is realized, and the effective segmentation of positive and negative terrains has important theoretical value and guiding significance for soil erosion control and ecological restoration and reconstruction in the Loess Plateau. [Methods] The typical sample area was selected in the hilly area of the Loess Plateau, and the terrain segmentation data set was made by using the medium resolution DEM data. The positive and negative terrains segmentation model of the improved Unet was constructed. Based on the Unet model structure, the residual module was introduced to replace the convolution module to deepen the network structure and increase the extraction of terrain information. Combined with the convolution attention module, eliminating useless information increased the anti-interference of the model; the activation function and loss function were optimized to enhance the robustness and accuracy of the model. [Results] The overall accuracy of the terrain segmentation of the slope deformation neighborhood judgment method is 70.3%. Among the deep learning models, the improved Unet model has the best effect, with a certain improvement compared with both the Unet model and the Res-Unet model, and the overall accuracy reaches 86.2%. [Conclusion] Compared with the traditional slope distortion neighborhood judgment method, the accuracy evaluation index of the network model segmentation results based on deep learning is better, which verifies the effectiveness of the improved Unet model in terrain segmentation.

References:: [1] 李锐.中国黄土高原研究与展望[M].北京:科学出版社,2008:1-3.
[2] 甘枝茂.黄土高原地貌与土壤侵蚀研究[M].西安:陕西人民出版社,1989:3-6.
[3] 许强,陈婉琳,蒲川豪,等.基于自然的解决方案在黄土高原重大工程灾变防控中的理论与实践[J].工程地质学报,2022,30(4):1179-1192.
[4] 穆兴民,李朋飞,刘斌涛,等.1901—2016年黄土高原土壤侵蚀格局演变及其驱动机制[J].人民黄河,2022,44(9):36-45.
[5] 刘迪,陈海,耿甜伟,等.基于地貌分区的陕西省区域生态风险时空演变[J].地理科学进展,2020,39(2):243-254.
[6] 汤国安,那嘉明,程维明.我国区域地貌数字地形分析研究进展[J].测绘学报,2017,46(10):1570-1591.
[7] 程维明,周成虎,申元村,等.中国近40年来地貌学研究的回顾与展望[J].地理学报,2017,72(5):755-775.
[8] 罗来兴.划分晋西、陕北、陇东黄土区域沟间地与沟谷地的地貌类型[J].地理学报,1956,22(3):201-222.
[9] 周毅.基于DEM的黄土高原正负地形及空间分异研究[D].南京:南京师范大学,2011.
[10] 张磊,汤国安,李发源,等.黄土地貌沟沿线研究综述[J].地理与地理信息科学,2012,28(6):44-48.
[11] Yang Xin, Li min, Na Jiaming, et al. Gully boundary extraction based on multidirectional hill-shading from high-resolution DEMs[J]. Transactions in Gis, 2017,21(6):1204-1216.
[12] Dai Wen, Yang Xin, Na Jiaming, et al. Effects of DEM resolution on the accuracy of gully maps in loess hilly areas[J]. Catena, 2019,177:114-125.
[13] Zhou Yi, Tang Guoan, Yang Xin, et al. Positive and negative terrains on northern Shaanxi Loess Plateau[J]. Journal of Geographical Sciences, 2010,20(1):64-76.
[14] 刘鹏举,朱清科,吴东亮,等.基于栅格DEM与水流路径的黄土区沟缘线自动提取技术研究[J].北京林业大学学报,2006,28(4):72-76.
[15] 王轲,王琤,张青峰,等.地形开度和差值图像阈值分割原理相结合的黄土高原沟沿线提取法[J].测绘学报,2015,44(1):67-75.
[16] Na Jiaming, Yang Xin, Dai Wen, et al. Bidirectional DEM relief shading method for extraction of gully shoulder line in loess tableland area[J]. Physical Geography, 2018,39(4):368-386.
[17] 晏实江,汤国安,李发源,等.利用DEM边缘检测进行黄土地貌沟沿线自动提取[J].武汉大学学报:信息科学版,2011,36(3):363-367.
[18] Jiang Sheng, Tang Guoan, Liu Kai. A new extraction method of loess shoulder-line based on Marr-Hildreth operator and terrain mask[J]. Plos One, 2015,10(4):e0123804.
[19] 周毅,汤国安,习羽,等.引入改进Snake模型的黄土地形沟沿线连接算法[J].武汉大学学报:信息科学版,2013,38(1):82-85.
[20] 宋效东,汤国安,周毅,等.基于并行GVF Snake模型的黄土地貌沟沿线提取[J].中国矿业大学学报,2013,42(1):134-140.
[21] 刘玮,李发源,熊礼阳,等.基于区域生长的黄土地貌沟沿线提取方法与试验[J].地球信息科学学报,2016,18(2):220-226.
[22] Ronneberger Olaf, Fischer Philipp, Brox Thomas. U-net:Convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2015:234-241.
[23] 许玥.基于改进Unet的遥感影像语义分割在地表水体变迁中的应用[D].重庆:重庆师范大学,2019.
[24] 卫佳杰.基于U-Net和注意力机制的双源遥感图像语义分割方法研究[D].西安:西安电子科技大学,2020.
[25] 徐佳伟,刘伟,单浩宇,等.基于PRCUnet的高分遥感影像建筑物提取[J].地球信息科学学报,2021,23(10):1838-1849.
[26] Cao Kaili, Zhang Xiaoli. An improved res-unet model for tree species classification using airborne high-resolution images[J]. Remote Sensing, 2020,12(7):1128.
[27] Hou Yuewu, Liu Zhaoying, Zhang Ting, et al. C-UNet:Complement UNet for Remote Sensing Road Extraction[J]. Sensors, 2021,21(6):2153.
[28] Nagi Anmol Sharan, Kumar Devinder, Sola Daniel, et al. RUF:Effective Sea Ice Floe Segmentation Using End-to-End RES-UNET-CRF with Dual Loss[J]. Remote Sensing, 2021,13(13):2460.
[29] 王祯,吴金华,白帅,等.黄土丘陵沟壑区耕地细碎化评价与土地整治工程分区:以吴起县为例[J].水土保持研究,2022,29(4):300-307.
[30] 刘长海,葛瑶,秦家凤,等.吴起县不同土地利用类型土壤物理性质特征[J].延安大学学报:自然科学版,2022,41(2):77-82.
[31] 王旭冉.基于改进U-Net的遥感影像语义分割研究[D].银川:宁夏大学,2020.
[32] 田萱,王亮,孟祥光.基于深度学习的图像语义分割技术[M].北京:海洋出版社,2019:86-87.
[33] Ramachandran P, Zoph B, Le Q V. Searching for activation functions[J]. Arxiv Preprint ArXiv:1710.05941, 2017, Doi:org/10.48550/arXiv.1710.05941.
[34] Howard Andrew, Sandler Mark, Chu Grace, et al. Searching for mobilenetv3[C]. Proceedings of the Ieee/Cvf International Conference on Computer Vision, 2019:1314-1324.
[35] Lin Tsung Yi, Goyal Priya, Girshick Ross, et al. Focal loss for dense object detection[C]. Proceedings of the Ieee International Conference on Computer Vision, 2017:2980-2988.
[36] Salehi Seyed Sadegh Mohseni, Erdogmus Deniz, Gholipour Ali. Tversky loss function for image segmentation using 3 D fully convolutional deep networks[C]. International Workshop on Machine Learning in Medical Imaging. Springer, Cham, 2017:379-387.
[37] Woo Sanghyun, Park Jongchan, Lee Joon Young, et al. Cbam:Convolutional block attention module[C]. Proceedings of the European Conference on Computer Vision(Eccv),2018:3-19.
[38] 王海涛,王一琛,王永强,等.基于MS-UNet的Landsat影像云检测[J].激光与光电子学进展,2021,58(14):87-94.
[39] Jiao Libin, Huo Lianzhi, Hu Changmiao, et al. Refined UNet:UNet-based refinement network for cloud and shadow precise segmentation[J]. Remote Sensing, 2020,12(12):2001.

Similar References:

Memo

Last Update: 2023-08-10