High-resolution remote sensing monitoring and evaluation engineering mode of soil erosion based on system engineering
LUO Zhidong, LIU Erjia, QI Shi, ZHAO Yuan
1. School of Soil and Water Conservation, Beijing Forestry University, 100083, Beijing, China;
2. The Center of Water and Soil Monitoring, Ministry of Water Resources, 100053, Beijing, China;
3. Forestry College, Beijing Forestry University, 100083, Beijing, China;
4. Forestry Ecological Engineering Research Center of Ministry of Education, 100083, Beijing, China
Abstract：[Background] High-resolution remote sensing technology is an important means to monitor and evaluate soil erosion. The regional applicability of high-resolution remote sensing information extraction automation algorithm is poor, the multi-period dynamic error fluctuation of remote sensing information extraction results is large, the integration of remote sensing information extraction and soil erosion monitoring and evaluation business is poor, and the overall application efficiency is not high.[Methods] Based on the research status of soil erosion monitoring and evaluation by high-resolution remote sensing, this paper considered the application of remote sensing technology in soil erosion monitoring and evaluation as a whole by using the principles and methods of system engineering, and put forward the construction idea of engineering model of high-resolution remote sensing monitoring and evaluation of soil erosion. This paper also clarified the conceptual connotation and main characteristics of the model, and put forward the overall framework of the model, as well as the construction content and implementation approach of the key elements of the model.[Results] After research, the model mainly includes 3 core elements:engineering comprehensive knowledge base model, engineering remote sensing information extraction algorithm model and engineering integrated model. Among them, the engineering comprehensive knowledge base model is mainly based on the spatial management framework system of parallel integration of soil and water conservation zoning and county administrative divisions, which standardizes and solidifies business and expert knowledge; the engineering remote sensing information extraction algorithm model is mainly to extract target objects and sub-regions for different high-resolution remote sensing information. An information extraction algorithm set with model and algorithm as the core is established, and the engineering integrated model mainly integrates the basic data, application software system and infrastructure equipment environment. In Hengshan county of Yulin city, Shanxi province, a demonstrating application on agriculture and animal husbandry was carried out. Automatic overall classification accuracy was 87.58%. After the engineering model corrected, classification accuracy was 95%. Based on the engineering application of the system platform and using integrated basic data, we completed the evaluation of soil erosion in Hengshan county, andthe overall efficiency of the evaluation increased by 2-3 times.[Conclusions] This model realizes systematic and collaborative optimization of monitoring and evaluating soil erosion using remote sensing, greatly improves its the stability, accuracy, practicability and comparability, and may provide technical support to promote dynamic assessment and governance of Chinese soil erosion.
赵辉,黎家作,李晶晶.中国水土流失动态监测与评价的现状与对策析[J].水土保持通报,2016,36(1):115. ZHAO Hui,LI Jiazuo,LI Jingjing. Present situation and countermeasures of dynamic monitoring and evaluation of soil and water loss in China[J]. Bulletin of Soil and Water Conservation,2016,36(1):115.
[2]
李智广,曹炜,刘秉正,等. 我国水土流失状况与发展趋势研究[J]. 中国水土保持科学,2008,6(1):57. LI Zhiguang,CAO Wei,LIU Bingzheng,et al. Current status and developing trend of soil erosion in China[J]. Science of Soil & Water Conservation,2008,6(1):57.
[3]
郭琳,裴志远,吴全,等. 面向对象的土地利用/覆盖遥感分类方法与流程应用[J]. 农业工程学报,2010,26(7):194. GUO Lin,PEI Zhiyuan,WU Quan,et al. Application of method and process of object-oriented land use-cover classification using remote sensing images[J]. Transactions of the CSAE,2010,26(7):194.
[4]
林卉,邵聪颖,李海涛,等. 高分辨率遥感影像5种面向对象分类方法对比研究[J]. 测绘通报,2017(11):17. LIN Hui,SHAO Congying,LI Haitao,et al. Five object-oriented classification methods analysis based on high-resolution remote sensing image[J]. Bulletin of Surveying and Mapping,2017(11):17.
[5]
张雁. 基于机器学习的遥感图像分类研究[D]. 北京:北京林业大学,2014:10. ZHANG Yan. Study of remote sensing image classification based on machine larning[D]. Beijing:Beijing Forestry University,2014:10.
[6]
杜培军,夏俊士,薛朝辉,等. 高光谱遥感影像分类研究进展[J]. 遥感学报,2016,20(2):236. DU Peijun,XIA Junshi,XUE Zhaohui,et al. Review of hyperspectral remote sensing image classification[J]. Journal of Remote Sensing,2016,20(2):236.
[7]
刘大伟,韩玲,韩晓勇. 基于深度学习的高分辨率遥感影像分类研究[J]. 光学学报,2016(4):298. LIU Dawei,HAN Ling, HAN Xiaoyong. High spatial resolution remote sensing image classification based on deep learning[J]. Acta Optica Sinica,2016(4):298.
[8]
汪应洛. 系统工程学[M]. 北京:高等教育出版社,2007:22. WANG Yingluo. Systems engineering[M]. Beijing:Higher Education Press,2007:22.
[9]
陈忠,盛毅华. 现代系统科学学[M]. 上海:上海科学技术文献出版社,2005. CHEN Zhong,SHEN Yihua. Modern system science[M]. Shanghai:Shanghai Scientific and Technological Literature Press,2005.
[10]
程理民. 运筹学模型与方法教程[M]. 北京:清华大学出版社,2000:89. CHENG Limin. Operations research models and methods tutorial[M]. Beijing:Tsinghua University Press,2000:89.
[11]
兰樟仁,张东水. 遥感影像多目标优化信息提取模式研究[J]. 农业工程学报,2008,24(7):155. LAN Zhangren,ZHANG Dongshui. Remote sensed information extracting model based on multiple-objective decision support theory:A case study of wetland agriculture[J]. Transactions of the CSAE,2008,24(7):155.
2019, Vol. 17 
