刍议服务于深度学习研究的林业遥感数据集建设

编号 lyqk009500

中文标题 刍议服务于深度学习研究的林业遥感数据集建设

作者 武红敢  王成波 

作者单位 1. 中国林业科学研究院资源信息研究所，北京 100091;
2. 国家林业和草原局林业遥感与信息技术实验室，北京 100091;
3. 中国科学院空天信息创新研究院，北京 100101

期刊名称 世界林业研究 

年份 2021 

卷号 34

期号 3

栏目名称 专题论述 

中文摘要 林业遥感数据集是林草行业开展深度学习研究的重要基础，更是林草资源实现自动化、智能化和动态化精准监测的重要保障。文中分析了自然图像和遥感图像数据集建设现状与方法，通过对深度学习算法在林业专题信息识别和分类中最新研究成果的阐述，表明建立林业遥感专用数据集的必要性和紧迫性；基于人工智能在图像处理领域的发展经验，提出林草业遥感影像数据集的建设思路和原则。

关键词 林草资源  精准监测  深度学习  遥感数据集  资源共享 

基金项目 “十三五”国家重点研发计划“野果林病虫害控制技术与监测、预警体系构建”（2016YFC0501503）

英文标题 Construction of Forestry Remote Sensing Dataset for Deep Learning

作者英文名 Wu Honggan, Wang Chengbo

单位英文名 1. Research Institute of Forest Resources Information Techniques, Chinese Academy of Forestry, Beijing 100091, China;
2. Laboratory of Forestry Remote Sensing and Information System, National Forestry and Grassland Administration, Beijing 100091, China;
3. Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, China

英文摘要 Forestry remote sensing dataset is the important basis for the forestry and grassland sector to carry out deep learning, and also the vital guarantee for the automatic, intelligent and dynamic monitoring of grassland and forest resources. This paper analyzes the status and methods of the natural images and remote sensing image dataset construction, and then expounds the necessity and urgency of building special forestry remote sensing dataset by introducing the latest research achievements in using the deep learning algorithms for special forestry information recognition and classification. At the end, the paper proposes the ideas and principles to construct the forestry and grassland remote sensing image dataset based on the development of artificial intelligence in image processing field.

英文关键词 forest and grass resources;accurate monitoring;deep learning;remote sensing dataset;resources sharing

起始页码 58

截止页码 62

投稿时间 2020/9/5

最后修改时间 2020/11/17

作者简介 武红敢，女，研究员，山西祁县人，1983年毕业于武汉测绘科技大学，主要从事遥感、地理信息系统、全球导航卫星系统在林业上的应用研究，E-mail：wuhg@ifrit.ac.cn

通讯作者介绍 王成波，男，博士，天津宝坻人，2010年毕业于武汉大学遥感信息工程学院，主要从事遥感、地理信息系统、人工智能在空间信息领域的应用研究，E-mail：wangcb@aircas.ac.cn

E-mail wuhg@ifrit.ac.cn;wangcb@aircas.ac.cn

分类号 S771.8

DOI 10.13348/j.cnki.sjlyyj.2020.0124.y

参考文献 [1] HINTON G E, OSINDERO S, TEH Y W. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18(7):1527-1554.
[2] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Image-net classification with deep Convolutional Neural Networks[C]. Neural Information Processing Systems, Harrahs and Harveys, Lake Tahoe, 2012.
[3] 刘环宇. 自动识别应用在互联网上崛起[EB/OL]. (2013-08-30)[2020-08-30]. http://www.cqn.com.cn/zj/content/2013-08/30/content_1928154.htm.
[4] LIN T Y, MAIRE M, BELONGIE S, et al. Microsoft coco: common objects in context[C]. European Conference on Computer Vision, Zurich, Switzerland, 2014.
[5] ALEX K. The CIFAR-10 dataset [EB/OL]. (2009-04-08) [2020-08-30]. http://www.cs.toronto.edu/~kriz/cifar.html.
[6] YANN L. The MNIST DATABASE of handwritten digits[EB/OL]. (1998-11-01)[2020-08-30]. http://yann.lecun.com/exdb/mnist/.
[7] YU F, ARI S, ZHANG Y D, et al. LSUN: construction of a large-scale image dataset using deep learning with humans in the loop[EB/OL]. (2015-6-10)[2020-08-30]. https://arxiv.org/abs/1506.03365.
[8] RUSSAKOVSKY O, DENG J, SU H, et al. Imagenet large scale visual recognition challenge[J]. International Journal of Computer Vision, 2015, 115(3):211-252.
[9] LIU Z W, LUO P, WANG X, et al. Large-scale CelebFaces Attributes (CelebA) dataset[EB/OL]. [2020-08-30]. http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html.
[10] SUMBUL G, CHARFUELAN M, DEMIR B, et al. BigEarthNet: a lrge-scale benchmark archive for remote sensing image understanding[C].IEEE International Geoscience and Remote Sensing Symposium, Yokohama, Japan, 2019: 5901-5904.
[11] SAIKAT B, SANGRAM G, SUPRATIK M, et al. DeepSat: a learning framework for satellite imagery[C]. 23rd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, Seattle, Washington, USA, 2015.
[12] XIA G, HU J, HU F, et al. Aid: a benchmark data set for performance evaluation of aerial scene classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(7):3965-3981.
[13] 赵理君, 郑柯, 史路路, 等. 全国地表类型遥感影像样本数据集[J]. 中国科学数据,2019,4(2):200-211. DOI:10.11922/csdata.2018.0058.zh
[14] ZHANG L, ZHANG L, DU B. Deep learning for remote sensing data: a technical tutorial on the state of the art[J]. IEEE Geoscience and Remote Sensing, 2016, 4(2):22-40.
[15] CHENG G, HAN J, lU X. Remote sensing image scene classification: benchmark and state of the art[J]. Proceedings of the IEEE, 2017, 105(10):1865-1883.
[16] ZHU X X, TUI A D. Deep learning in remote sensing: a comprehensive review and list of resources[J]. IEEE Geoscience and Remote Sensing, 2017, 5(42):8-36.
[17] 李海涛, 戴莉莉, 顾海燕, 等. 样本尺寸对遥感影像FCN训练模型的影响分析[J]. 测绘科学,2019,44(6):133-137.
[18] REAl E, AGGARWAL A, HUANG Y P, et al. Regularied evolution for image classifier architecture search[C]. The Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19), Honolulu, Hawaii, 2019.
[19] 王敏, 陈金勇, 王港, 等. 面向深度学习的遥感图像目标样本生成方法[J]. 国外电子测量技术,2019,38(4):60-65.
[20] 程滔, 吴芸, 郑新燕, 等. 面向遥感影像智能分类的海量样本数据采集方法[J]. 测绘通报,2019(10):56-60.
[21] 杨会贇. 基于深度学习的高分遥感影像城镇森林信息提取研究[D]. 北京: 中国地质大学(北京), 2019.
[22] 王莉影, 高心丹. 基于迁移学习的遥感影像树种类型分类[J]. 实验室研究与探索,2019,38(9):29-33.
[23] 陈厚坤, 孙仁诚. 基于深度学习的遥感图像茶园区域识别应用研究[J]. 工业控制计算机,2020,33(2):93-96.
[24] 郭颖, 李增元, 陈尔学, 等. 一种改进的高空间分辨率遥感影像森林类型深度学习精细分类方法: 双支FCN-8s[J]. 林业科学,2020,56(3):48-60.
[25] 殷存军. 基于深度卷积网络和AdaBoost算法的无人机图像中病害松树识别[D]. 合肥: 安徽大学, 2020.
[26] SLADOJEVIC S, ARSENOVIC M, NDERLA A, et al. Deep neural networks based recognition of plant diseases by leaf image classification[J/OL]. Computational Intelligence and Neuroscience, 2016: 3289801. DOI: 10.1155/2016/3289801.
[27] WEGNERA J D, BRANSON S, HALL D, et al. Cataloging public objects using aerial and street-level images-urban trees[C]. 2016 IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, 2016.
[28] TOMOHIRO M, AKIRA I, HIROYUKI N, et al. Lidar-based individual tree species classification using convolutional neural network[C]. SPIE Optocal Metrology, Munich, 2017.
[29] LI W J, FU H H, YU L, et al. Deep learning based oil palm tree detection and counting for high-resolution remote sensing images[J]. Remote Sensing, 2016, 9(1):22. DOI:10.3390/rs9010022
[30] LI W J, FU H H, YU L. Deep convolutional neural network based large-scale oil palm tree detection for high-resolution remote sensing images[C]. 2017 IEEE International Geoscience and Remote Sensing Symposium(IGARSS), Fort Worth, TX, 2017.
[31] LI W J, DONG R M, FU H H, et al. Large-scale oil palm tree detection from high-resolution satellite images using two-stage convolutional neural networks[J]. Remote Sensing, 2019, 11(1):11-30.
[32] QIAO R, GHODSI A, WU H G, et al. Simple weakly supervised deep learning pipeline for detecting individual red-attacked trees in VHR remote sensing images[J]. Remote Sensing Letters, 2020, 11(7):650-658.
[33] 黄华国. 林业定量遥感研究进展和展望[J]. 北京林业大学学报,2019,41(12):1-14.
[34] 孙熠, 李培军. 利用主动学习改进遥感图像单类分类: 以正类和未标记样本学习方法为例[J]. 北京大学学报(自然科学版),2020,56(1):155-163.
[35] 胡龙廷, 张克. 深度学习方法用于遥感图像处理的研究进展[J]. 计算机产品与流通,2019(6):234.

发布日期 2020-12-01

PDF全文 浏览全文