深度学习遥感影像解译技术在耕地保护中的应用

doi:10.13474/j.cnki.11-2246.2023.0247

摘要/Abstract

摘要： 耕地保护事关国家粮食安全、生态安全和社会稳定,是国计民生的头等大事。深度学习技术在海量数据分析领域的广泛应用,为高效、精准的遥感影像解译提供了技术基础。本文研究了基于深度学习的遥感影像解译技术,利用遥感影像数据和对应的矢量数据构建了可供训练的解译样本库,并提出了一种基于深度残差网络结构的解译模型。通过试验证明了该模型的实用性,实现了对试验区域主要地类面积的变化监测,对比多期影像解译结果和增减挂钩等业务数据,验证了耕地复垦的实施情况。结果表明,深度学习遥感影像解译技术在耕地保护领域有较广泛的应用前景。

关键词: 耕地保护, 深度学习, 高分辨率遥感影像

Abstract: Cultivated land protection is related to national food security, ecological security and social stability. The wide application of deep learning technology in the field of massive data analysis provides a technical foundation for efficient and accurate remote sensing image interpretation. In this paper, the remote sensing image interpretation technology based on deep learning is studied, and a sample library of interpretation samples for training is constructed by using remote sensing image data and corresponding vector data, and an interpretation model based on deep residual network structure is proposed. The practicality of the model has been proved by experiments. In this paper, the change of the main land area in the experimental area is monitored, and the implementation of cultivated land reclamation is verified by comparing the multi-phase image interpretation results and the business data such as the increase and decrease linkage. The results show that deep learning remote sensing image interpretation technology has a wide application prospect in the field of cultivated land protection.

Key words: cultivated land protection, deep learning, high-resolution remote sensing image

中图分类号:

P237

张戬, 高雅. 深度学习遥感影像解译技术在耕地保护中的应用[J]. 测绘通报, 2023, 0(8): 142-145.

ZHANG Jian, GAO Ya. Application of deep learning remote sensing image interpretation technology in cultivated land protection[J]. Bulletin of Surveying and Mapping, 2023, 0(8): 142-145.

参考文献

[1] 黄瑾,陈红霞. 城乡建设用地增减挂钩政策评价——基于文献综述[J]. 中国国土资源经济,2021,34(9):82-89.
[2] 张良培,武辰. 多时相遥感影像变化检测的现状与展望[J]. 测绘学报,2017,46(10):1447-1459.
[3] 高峰,楚博策,帅通,等. 基于深度学习的耕地变化检测技术[J]. 无线电工程,2019,49(7):571-574.
[4] 吕盛坪,李灯辉,冼荣亨. 深度学习在我国农业中的应用研究现状[J]. 计算机工程与应用,2019,55(20):24-33.
[5] LECUN Y,BENGIO Y,HINTON G. Deep learning[J]. Nature,2015,521(7553):436-444.
[6] LECUN Y,BOTTOU L,ORR G B,et al. Efficient BackProp[M]//ORR G B,MÜLLER K R,Eds. Lecture Notes in Computer Science. Berlin,Heidelberg:Springer Berlin Heidelberg,1998:9-50.
[7] RUDER S. An overview of gradient descent optimization algorithms[EB/OL]. 2016-09-15[2022-09-22].https://arxiv.org/abs/1609.04747.
[8] SHELHAMER E,LONG J,DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(4):640-651.
[9] KRIZHEVSKY A,SUTSKEVER I,HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM,2017,60(6):84-90.
[10] SZEGEDY C,LIU Wei,JIA Yangqing,et al. Going deeper with convolutions[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition.Boston:IEEE,2015:1-9.
[11] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al. Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas:IEEE,2016:770-778.
[12] TONG Xinyi,XIA Guisong,LU Qikai,et al. Land-cover classification with high-resolution remote sensing images using transferable deep models[EB/OL]. 2018-07-16[2022-09-22]. https://arxiv.org/abs/1807.05713.
[13] WANG Guoli,FAN Bin,XIANG Shiming,et al. Aggregating rich hierarchical features for scene classification in remote sensing imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2017,10(9):4104-4115.
[14] HUANG Bo,ZHAO Bei,SONG Yimeng. Urban land-use mapping using a deep convolutional neural network with high spatial resolution multispectral remote sensing imagery[J]. Remote Sensing of Environment,2018,214:73-86.
[15] CHEN L C,PAPANDREOU G,SCHROFF F,et al. Rethinking atrous convolution for semantic image segmentation[EB/OL]. 2017-06-17[2022-09-21]. https://arxiv.org/abs/1706.05587.