[1]毛星,金晶,张欣,等.基于改进DeepLab v3+模型和迁移学习的高分遥感耕地信息提取方法[J].江苏农业学报,2023,(07):1519-1529.[doi:doi:10.3969/j.issn.1000-4440.2023.07.009]
　MAO Xing,JIN Jing,ZHANG Xin,et al.High-resolution remote sensing arable land information extraction method based on improved DeepLab v3+ model and transfer learning[J].,2023,(07):1519-1529.[doi:doi:10.3969/j.issn.1000-4440.2023.07.009]
点击复制

基于改进DeepLab v3+模型和迁移学习的高分遥感耕地信息提取方法()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2023年07期
页码:
1519-1529
栏目:
农业信息工程
出版日期:
2023-10-31

文章信息/Info

Title:
High-resolution remote sensing arable land information extraction method based on improved DeepLab v3+ model and transfer learning
作者:
毛星1; 2; 金晶1; 2; 张欣1; 2; 戴佩玉1; 2; 任妮1; 2
(1.江苏省农业科学院农业信息研究所，江苏南京210014；2.农业农村部长三角智慧农业技术重点实验室，江苏南京210014)
Author(s):
MAO Xing1; 2; JIN Jing1; 2; ZHANG Xin1; 2; DAI Pei-yu1; 2; REN Ni1; 2
(1.Institute of Agricultural Information, Jiangsu Academy of Agricultural Sciences, Nanjing 210014, China;2.Key Laboratory of Intelligent Agricultural Technology (Changjiang Delta), Ministry of Agriculture and Rural Affairs, Nanjing 210014, China)
关键词:
耕地信息提取; 迁移学习; DEA-Net; 高分遥感
Keywords:
arable land information extraction; transfer learning; DEA-Net; high resolution remote sensing
分类号:
S126
DOI:
doi:10.3969/j.issn.1000-4440.2023.07.009
文献标志码:
A
摘要:
针对实际遥感耕地信息提取工作中，多源数据特征复杂、样本标注工作繁重等导致高空间分辨率影像解译精度不高、自动化程度不够的问题，本研究基于DeepLab v3+模型，提出一种融合邻域边缘加权模块(NEWM)和轴向注意力机制模块(CBAM-s)的卷积网络模型DEA-Net，结合迁移学习方法进行高分辨率遥感影像耕地信息提取。首先，在浅层网络结构中加入邻域边缘加权模块，提升高分辨率下地物的连续性，细化边缘分割粒度；其次，在深层网络结构中添加轴向注意力机制模块，增加细小地物的关注权重，减少深度卷积导致地物丢失的情况；最后，采用迁移学习的思想，降低样本标注工作量，提高模型学习能力。利用高分卫星土地覆盖数据集(GID)数据构建源域数据集进行模型预训练，将获取的模型参数及权重信息迁移至大数据与计算智能大赛(BDCI)遥感影像地块分割竞赛数据集和全国人工智能大赛(NAIC)遥感影像数据集制作的2种不同目标域数据集中，微调训练后应用于耕地信息提取研究。结果表明，本研究构建方法能够增强模型的空间细节学习能力，提高耕地语义分割精度的同时，降低2/3以上的训练样本数量，为遥感耕地信息提取及农业数据智能化利用提供新的思路和方法。
Abstract:
The complex multi-source data features and heavy sample annotation work in the practical remote sensing arable land information extraction work will lead to low accuracy and insufficient automation of high spatial resolution image interpretation. In view of the above problems, based on DeepLab v3+, we proposed a convolutional network model DEA-Net that incorporated the neighborhood edge weighting module (NEWM) and the axial attention mechanism (CBAM-s), and combined the transfer learning method to extract arable land information of high-resolution remote sensing images. First, the NEWM was added to the shallow network structure to improve the continuity of features under high resolution and refine the granularity of edge segmentation. Then, the CBAM-s was added to the deep network structure to increase the attention weight of fine features and reduce the loss of features due to deep convolution. Finally, the idea of transfer learning was adopted to reduce the sample annotation workload and improve the learning ability of the model. The source domain dataset was constructed using the Gaofen image dataset (GID) for model pre-training, and the acquired model parameters and weight information were migrated to two different target domain datasets produced by big data & computing intelligence contest (BDCI) and national artificial intelligence challenge (NAIC), and fine-tuned and trained for arable land information extraction. The results showed that the method constructed in this study could enhance the spatial detail learning ability of the model, improve the semantic segmentation accuracy of arable land, and reduce the number of training samples by more than 2/3. It can provide new ideas and methods for remote sensing arable land information extraction and intelligent utilization of agricultural data.

参考文献/References:

[1]唐华俊. 农业遥感研究进展与展望[J]. 农学学报， 2018, 8(1): 167-171.
[2]冯权泷，牛博文，朱德海，等. 土地利用/覆被深度学习遥感分类研究综述[J]. 农业机械学报， 2022, 53(3): 1-17.
[3]ARANGO R B, DAZ I, CAMPOS A, et al. Automatic arable land detection with supervised machine learning[J]. Earth Science Informatics, 2016, 9: 535-545.
[4]陈杰，陈铁桥，刘慧敏，等. 高分辨率遥感影像耕地分层提取方法[J]. 农业工程学报, 2015, 31(3): 190-198.
[5]朱江洪，李江风，叶菁. 利用决策树工具的土地利用类型遥感识别方法研究[J]. 武汉大学学报(信息科学版), 2011, 36(3): 301-305.
[6]NITZE I, BARRETT B, CAWKWELL F. Temporal optimisation of image acquisition for land cover classification with random forest and MODIS time-series[J]. International Journal of Applied Earth Observation and Geoinformation, 2015, 34: 136-146.
[7]LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[C]. New York: IEEE Press, 2015.
[8]RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation[C]. Munich: Springer International Publishing, 2015.
[9]CHEN L, 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.
[10]YAO X, GUO Q, LI A. Light-weight cloud detection network for optical remote sensing images with attention-based DeepLab v3+ architecture[J]. Remote Sensing, 2021, 13. DOI:org/10.3390/rs13183617.
[11]WANG Z, WANG J, YANG K, et al. Semantic segmentation of high-resolution remote sensing images based on a class feature attention mechanism fused with DeepLab v3+[J]. Computers & Geosciences, 2022, 158. DOI:org/10.1016/j.cageo.2021.104969.
[12]刘文祥，舒远仲，唐小敏，等. 采用双注意力机制 DeepLab v3+算法的遥感影像语义分割[J]. 热带地理, 2020, 40(2): 303-313.
[13]叶沅鑫，谭鑫，孙苗苗，等. 基于增强 DeepLab v3网络的高分辨率遥感影像分类[J]. 测绘通报，2021(4): 40-44.
[14]WANG Y, GAO L, HONG D, et al. Mask DeepLab: end-to-end image segmentation for change detection in high-resolution remote sensing images[J]. International Journal of Applied Earth Observation and Geoinformation, 2021, 104. DOI:org/10.1016/j.jag.2021.102582.
[15]GUO H, DU B, ZHANG L, et al. A coarse-to-fine boundary refinement network for building footprint extraction from remote sensing imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2022, 183: 240-252.
[16]HUANG J, ZHANG X, SUN Y, et al. Attention-guided label refinement network for semantic segmentation of very high resolution aerial orthoimages[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 4490-4503.
[17]ZHU Q, LIAO C, HU H, et al. MAP-Net: multiple attending path neural network for building footprint extraction from remote sensed imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 59(7): 6169-6181.
[18]WOO S, PARK J, LEE J Y, et al. CBAM : convolutional block attention module[C]. Munich: Springer International Publishing, 2018.
[19]WANG D, CHEN X, JIANG M, et al. ADS-Net: an attention-based deeply supervised network for remote sensing image change detection[J]. International Journal of Applied Earth Observation and Geoinformation, 2021, 101. DOI:org/10.1016/j.jag.2021.102348.
[20]RIBANI R, MARENGONI M. A survey of transfer learning for convolutional neural networks[C]. Rio de Janeiro: IEEE, 2019.

相似文献/References:

[1]宋晓倩,张学艺,张春梅,等.基于深度迁移学习的酿酒葡萄种植信息提取[J].江苏农业学报,2020,(03):689.[doi:doi:10.3969/j.issn.1000-4440.2020.03.022]
　SONG Xiao-qian,ZHANG Xue-yi,ZHANG Chun-mei,et al.Extraction of wine grape planting information based on deep transfer learning[J].,2020,(07):689.[doi:doi:10.3969/j.issn.1000-4440.2020.03.022]
[2]刘翱宇,吴云志,朱小宁,等.基于深度残差网络的玉米病害识别[J].江苏农业学报,2021,(01):67.[doi:doi:10.3969/j.issn.1000-4440.2021.01.009]
　LIU Ao-yu,WU Yun-zhi,ZHU Xiao-ning,et al.Corn disease recognition based on deep residual network[J].,2021,(07):67.[doi:doi:10.3969/j.issn.1000-4440.2021.01.009]
[3]牛学德,高丙朋,南新元,等.基于改进DenseNet卷积神经网络的番茄叶片病害检测[J].江苏农业学报,2022,38(01):129.[doi:doi:10.3969/j.issn.1000-4440.2022.01.015]
　NIU Xue-de,GAO Bing-peng,NAN Xin-yuan,et al.Detection of tomato leaf disease based on improved DenseNet convolutional neural network[J].,2022,38(07):129.[doi:doi:10.3969/j.issn.1000-4440.2022.01.015]
[4]周维,牛永真,王亚炜,等.基于改进的YOLOv4-GhostNet水稻病虫害识别方法[J].江苏农业学报,2022,38(03):685.[doi:doi:10.3969/j.issn.1000-4440.2022.03.014]
　ZHOU Wei,NIU Yong-zhen,WANG Ya-wei,et al.Rice pests and diseases identification method based on improved YOLOv4-GhostNet[J].,2022,38(07):685.[doi:doi:10.3969/j.issn.1000-4440.2022.03.014]

备注/Memo

备注/Memo:
收稿日期：2022-09-15基金项目：高分辨率对地观测系统重大专项(74-Y50G12-90-01-22/23)作者简介：毛星(1991-)，男，江苏镇江人，硕士，工程师，主要从事农业遥感数据分析与利用研究。(E-mail)maoxing@jaas.ac.cn通讯作者：任妮，(E-mail)rn@jaas.ac.cn
更新日期/Last Update: 2023-11-17