[1]方 炫,张同亮,彭明国,等.利用空间特征的平原地区灌溉沟渠优化提取[J].常州大学学报(自然科学版),2023,35(04):43-51.[doi:10.3969/j.issn.2095-0411.2023.04.007]
 FANG Xuan,ZHANG Tongliang,PENG Mingguo,et al.Optimal extraction method of irrigation ditches in plain areas using spatial characteristics[J].Journal of Changzhou University(Natural Science Edition),2023,35(04):43-51.[doi:10.3969/j.issn.2095-0411.2023.04.007]
点击复制

利用空间特征的平原地区灌溉沟渠优化提取()
分享到:

常州大学学报(自然科学版)[ISSN:2095-0411/CN:32-1822/N]

卷:
第35卷
期数:
2023年04期
页码:
43-51
栏目:
环境科学与工程
出版日期:
2023-07-28

文章信息/Info

Title:
Optimal extraction method of irrigation ditches in plain areas using spatial characteristics
文章编号:
2095-0411(2023)04-0043-09
作者:
方 炫1 张同亮2 彭明国1 蒋 圣3 张秋亚1
(1.常州大学 城市建设学院, 江苏 常州 213164; 2.泗阳县水利局, 江苏 宿迁 223700; 3.无锡工艺职业技术学院 机电与信息工程学院, 江苏 无锡 214206)
Author(s):
FANG Xuan1 ZHANG Tongliang2 PENG Mingguo1 JIANG Sheng3 ZHANG Qiuya1
(1.School of Urban Construction, Changzhou University, Changzhou 213164, China; 2.Siyang County Water Conservancy Bureau, Suqian 223700, China; 3.School of Mechatronics and Information, Wuxi Vocational Institute of Arts and Technology, Wuxi 214206, China)
关键词:
灌溉沟渠 面向对象法 无人机影像 空间特征 最近邻分类
Keywords:
irrigation ditches object-oriented method unmanned aerial vehicle imagery spatial characteristics nearest neighbor classification
分类号:
X 87
DOI:
10.3969/j.issn.2095-0411.2023.04.007
文献标志码:
A
摘要:
针对异物同谱和植被覆盖影响沟渠提取效果的问题,提出一种基于无人机影像和空间特征对提取过程进行优化和补偿的灌溉沟渠提取方法。首先,采用尺度参数估计(Estimation of Scale Parameter,ESP)方法确定最优分割参数; 其次,通过光谱均值及标准差构建特征空间,使用最近邻方法获得初始分类结果; 最后,利用长宽比、相对边界指标、边界规整度、面积等空间几何和关系特征,对初始分类结果进行优化提取。试验发现,对于空间分辨率为0.07 m的平原灌区影像,最优分割参数为61,基于空间特征的优化提取方法将提取精度提高到92.3%,较优化前提高了7.6%。该方法能有效解决杂草及异物同谱所导致的沟渠缺失及部分断连问题,提高平原地区灌溉沟渠提取的精度。
Abstract:
The capability of applying remote sensing data on ditch extraction is reduced because of vegetation coverage and different objects with same spectrum. To solve the problem, an integrated framework based on unmanned aerial vehicle imagery using spatial features to optimize and compensate the extraction process is proposed for irrigation ditch extraction. Firstly, the optimized segmentation parameter was achieved using the estimation of scale parameter algorithm(ESP). Secondly, the feature spaces for classification were constructed by spectral mean and standard deviation, and the initial classification was conducted by using the supervised k-nearest neighborhood classifier. Finally, the spatial geometric and relational features including aspect ratio, relative boundary index, boundary regularity and area, were used to optimize the initial classification results. The result shows that the optimal segmentation parameter is 61 for the plain irrigation area with 0.07 m resolution, and the optimization extraction method based on spatial features improves the extraction accuracy to 92.3%, which is 7.6% higher than that before optimization. The proposed method can effectively solve the problems of ditch missing and partial disconnection caused by the same spectrum of weeds and different objects with same spectrum, improving the accuracy of ditch extraction in plain irrigation areas.

参考文献/References:

[1] NAGARAJ D, PROUST E, TODESCHINI A, et al. A new dataset of global irrigation areas from 2001 to 2015[J]. Advances in Water Resources, 2021, 152: 103910.
[2] JÄGERMEYR J, PASTOR A, BIEMANS H, et al. Reconciling irrigated food production with environmental flows for sustainable development goals implementation[J]. Nature Communications, 2017, 8: 15900.
[3] CHAO Z H, FANG X, NA J M, et al. A collaborative sensing system for farmland water conservancy project maintenance through integrating satellite, aerial, and ground observations[J]. Water, 2021, 13(16): 2163.
[4] 李健, 徐旭. 基于无人机影像的煤场工程量监测分析[J]. 常州大学学报(自然科学版), 2019, 31(2): 35-43.
[5] 韩文霆, 郭聪聪, 张立元, 等. 基于无人机遥感的灌区土地利用与覆被分类方法[J]. 农业机械学报, 2016, 47(11): 270-277.
[6] 韩文霆, 张立元, 张海鑫, 等. 基于无人机遥感与面向对象法的田间渠系分布信息提取[J]. 农业机械学报, 2017, 48(3): 205-214.
[7] 张自超, 韩宇, 陈建, 等. 基于无人机遥感数据的生态渠系信息提取[J]. 排灌机械工程学报, 2018, 36(10): 1006-1011.
[8] 张宏鸣, 王斌, 韩文霆, 等. 基于全卷积神经网络的灌区无人机正射影像渠系提取[J]. 农业机械学报, 2019, 50(6): 241-248.
[9] CAZORZI F, FONTANA G D, DE LUCA A, et al. Drainage network detection and assessment of network storage capacity in agrarian landscape[J]. Hydrological Processes, 2013, 27(4): 541-553.
[10] SAH R K, DAS A K. Drainage skeletonization from flow-accumulated area without the use of threshold[J]. Geocarto International, 2015, 30(8): 894-904.
[11] 张宏鸣, 任强, 韩文霆, 等. 基于SVM的灌区无人机影像渠系提取[J]. 农业机械学报, 2018, 49(2): 141-148.
[12] 顾铮鸣, 金晓斌, 杨晓艳, 等. 基于无人机遥感影像监测土地整治项目道路沟渠利用情况[J]. 农业工程学报, 2018, 34(23): 85-93.
[13] DRGUT L, TIEDE D, LEVICK S R. A tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data[J]. International Journal of Geographical Information Science, 2010, 24(6): 859-871.
[14] LOURENCO P, TEODORO A C, GONCALVES J A, et al. Assessing the performance of different OBIA software approaches for mapping invasive alien plants along roads with remote sensing data[J]. International Journal of Applied Earth Observation and Geoinformation, 2021, 95: 102263.
[15] BAATZ M. Multiresolution segmentation: an optimization approach for high quality multi-scale image segmentation[J]. Angewandte Geographische Informations-verarbeitung, 2000, 22(3): 12-23.
[16] 刘凯. 基于多源数据的黄土高原(重点流失区)侵蚀沟提取及区域差异性研究[D]. 南京: 南京师范大学, 2017.
(责任编辑:李艳,周安迪)

备注/Memo

备注/Memo:
收稿日期: 2022-12-14。
基金项目: 国家自然科学基金资助项目(41871313); 江苏省自然科学基金面上资助项目(BK20161118)。
作者简介: 方炫(1982—), 女, 江苏沭阳人, 博士, 副教授。 通信联系人: 张同亮(1985—), E-mail: 2990244249@qq.com
更新日期/Last Update: 1900-01-01