Please wait a minute...
img

官方微信

遥感技术与应用  2023, Vol. 38 Issue (5): 1126-1135    DOI: 10.11873/j.issn.1004-0323.2023.5.1126
遥感应用     
基于孤立森林的水体异常快速发现与识别
朱秀芳1,2(),李原3,4,郭锐1,2
1.北京师范大学遥感科学国家重点实验室,北京 100875
2.北京师范大学 地理科学学部遥感科学与工程研究院,北京 100875
3.内蒙古大学 生态与环境学院,内蒙古 呼和浩特 010021
4.内蒙古自治区河流与湖泊生态重点实验室,内蒙古 呼和浩特 010021
Rapid Detection and Identification of Water Anomalies based on Isolated Forest
Xiufang ZHU1,2(),Yuan LI3,4,Rui GUO1,2
1.State Key Laboratory of Remote Sensing Science,Beijing Normal University,Beijing 100875,China
2.Institute of Remote Sensing Science and Engineering,School of Geographic Sciences,Beijing Normal University,Beijing 100875,China
3.College of Ecology and Environment,Inner Mongolia University,Hohhot 010021,China
4.Key Laboratory of River and Lake Ecology,Inner Mongolia Autonomous Region,Hohhot 010021,China
 全文: PDF(5739 KB)   HTML
摘要:

现有异常水体的检测研究通常针对特定区域、特定数据源和特定时相开展,且往往是事后的检测而非实时的监测,不能很好地服务于水体异常快速发现与识别的业务需求。为此,提出了一种基于无监督的孤立森林加决策规则(U-IForest-SD)的水体异常信息提取方法,并以Landsat与Sentinel的数据作为测试数据,以青岛浒苔、松雅湖黑臭水、墨西哥湾溢油为案例,对比了支持向量机、监督孤立森林以及U-IForest-SD 3种方法识别浒苔、黑臭水和溢油的精度。研究结果显示:该方法对于3种异常类型的总体识别精度都在90%以上、Kppa系数都在0.8以上,整体精度高于监督孤立森林但略低于SVM。该算法只需要输入单期影像,无需训练样本,具有可移植性好、普适性强、自动化程度高的优点。此外,该方法可以有效地避免“假警”和“虚警”的发生,在异常水体的快速发现和识别业务中有很好的应用前景。

关键词: 异常检测溢油黑臭水浒苔    
Abstract:

The existing detection research of abnormal water bodies is usually carried out for specific regions, specific data sources and specific time phases. Anomaly recognition algorithm testing is often a backtracking of the water body anomaly events that have occurred, rather than real-time monitoring of the anomaly events, which cannot serve the requirements of rapid detection and identification of water body anomalies. In this paper, a method of extracting water body abnormal information based on unsupervised isolated forest plus decision rule (U-IForest-SD) is proposed. We selected Landsat and Sentinel as the test data, and tested the accuracy of U-IForest-SD with the black and smelly water body of Qingdao Enteromorpha, Songya lake and the oil spill in the Gulf of Mexico as research cases. We also compared U-IForest-SD with SVM and supervised isolated forests. The results show that the overall accuracy of the proposed method for the three types of anomalies is above 90%, and the kappa coefficient is above 0.8. The overall accuracy is higher than that of supervised isolated forest but slightly lower than that of SVM. This algorithm only needs to input single phase images, and does not need training samples. It has the advantages of good portability, strong universality and high automation. In addition, it can effectively avoid the occurrence of "wrong alarm" and "false alarm". Therefore, the newly proposed method has a good application prospect in the rapid detection and identification of abnormal water bodies.

Key words: Anomaly detection    Oil spill    Black smelly water    Enteromorpha
收稿日期: 2022-09-12 出版日期: 2023-11-07
ZTFLH:  X87  
基金资助: 国家自然科学基金重大项目(41292583)
作者简介: 朱秀芳(1982-),女,浙江天台人,教授,主要从事遥感应用研究。E?mail:zhuxiufang@bnu.edu.cn
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
朱秀芳
李原
郭锐

引用本文:

朱秀芳,李原,郭锐. 基于孤立森林的水体异常快速发现与识别[J]. 遥感技术与应用, 2023, 38(5): 1126-1135.

Xiufang ZHU,Yuan LI,Rui GUO. Rapid Detection and Identification of Water Anomalies based on Isolated Forest. Remote Sensing Technology and Application, 2023, 38(5): 1126-1135.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2023.5.1126        http://www.rsta.ac.cn/CN/Y2023/V38/I5/1126

图1  研究区概况(a) 青岛浒苔 (b) 松雅湖黑臭水 (c) 墨西哥湾溢油
数据类型获取时间数据描述用途
Landsat 82021-07-09青岛浒苔发生时的数据测试算法提取浒苔的适用性
Landsat 82021-09-11青岛浒苔结束后的数据测试算法在正常水体中是否会错误识别异常
Landsat 82016-11-28松雅湖黑臭水体发生时的数据测试算法提取黑臭水体的适用性
Sentinel-22020-05-02墨西哥湾溢油发生时的数据测试算法提取溢油的适用性
表 1  数据详情
图2  正常水体与异常水体的对比(a) 反射率对比 (b) NDVI区间
类型NDVI
最小值Min最大值Max均值Mean标准差σ

范围

(Mean±2σ)

浒苔0.160.610.4080.13(0.15,0.67)
黑臭水体-0.303-0.13-0.2040.049(-0.302,-0.106)
溢油-0.980-0.880.23(-1,-0.42)
表 2  异常水体NDVI统计
图3  水体异常发现和识别的流程
图4  基于SVM、S-IForest和U-IForest 3类方法的异常水体识别结果
案例方法PAUAOAKappa
青岛浒苔SVM0.940.990.970.95
S-IForest0.840.900.900.79
U-IForest- IQR0.760.950.880.75
U-IForest- 1.5IQR0.760.960.890.77
U-IForest- SD0.780.990.900.80
U-IForest- 2SD0.720.960.870.72
UPCA-U-IForest-SD0.800.960.900.80

松雅湖

黑臭水

SVM0.990.990.990.99
S-IForest0.990.850.940.87
U-IForest- IQR0.430.860.800.46
U-IForest- 1.5IQR0.040.450.670.02
U-IForest- SD0.970.970.960.90
U-IForest- 2SD0.080.550.680.07
UPCA-U-IForest-SD0.990.930.970.95
墨西哥湾溢油SVM0.910.990.980.94
S-IForest0.470.970.860.56
U-IForest- IQR0.340.860.770.38
U-IForest- 1.5IQR0.110.970.800.17
U-IForest- SD0.840.950.940.85
U-IForest- 2SD0.330.860.770.37
UPCA-U-IForest-SD0.800.760.880.70
表3  精度验证
图5  仅依赖决策规则的异常水体识别结果
图6  仅使用U-IForest的异常水体识别结果
1 BLONDEAU-PATISSIER D, GOWER J F, DEKKER A G, et al. A review of ocean color remote sensing methods and statistical techniques for the detection, mapping and analysis of phytoplankton blooms in coastal and open oceans[J]. Progress in Oceanography, 2014(123): 123-144.
2 WANG Xiaoqin, CHEN Chongcheng. Application of remote sensing to environmental monitoring in coastal waters[J]. Marine Environmental Science,2000,19(4):72-76.
2 汪小钦,陈崇成.遥感在近岸海洋环境监测中的应用[J].海洋环境科学,2000,19(4):72-76.
3 LIN Xiaochen, YANG Liangliang. Applications of satellite-based remote sensing techniques in water pollution monitoring[J]. Sichuan Environment, 2023, 42(1): 306-314
3 林晓昇,杨亮亮.卫星影像遥感技术在水污染监测的应用[J].四川环境, 2023, 42(1):309-314.
4 LOU Xiulin, HUANG Weigen. An artificial neural network method for detecting red tides with NOAA AVHRR imagery[J]. Journal of Remote Sensing, 2003,8(2):125-130,162.
4 楼琇林,黄韦艮.基于人工神经网络的赤潮卫星遥感方法研究[J].遥感学报,2003,8(2):125-130,162.
5 MICHELI F. Derivation of Red Tide index and density using Geostationary Ocean Color Imager (GOCI) data[J]. Remote Sensing, 2021, 13.DOI:10.3390/rs13020298
doi: 10.3390/rs13020298
6 JIANG Dejuan, WANG Kun, XIA Yun. Comparative studies on remote sensing techniques for red tide monitoring in Bohai Sea[J]. Marine Environmental Science,2020,39(3):460-467.
6 姜德娟,王昆,夏云.渤海赤潮遥感监测方法比较研究[J].海洋环境科学,2020,39(3):460-467.
7 HU C, MULLER-KARGER F E, TAYLOR C J, et al. Red tide detection and tracing using modis fluorescence data: A regional example in SW Florida coastal waters[J]. Remote Sensing of Environment,2005,97(3):311-321. DOI:10. 1016/j.rse.2005.05.013
doi: 10. 1016/j.rse.2005.05.013
8 YE Na, JIA Jianjun, TIAN Jing, et al. Advances in the study of ulvapolifera monitoring with remote sensing[J]. Remote Sensing for Land&Resources,2013,25(1):7-12.
8 叶娜,贾建军,田静 等.浒苔遥感监测方法的研究进展[J].国土资源遥感,2013,25(1):7-12.
9 WANG Yiren, WANG Shengqiang, YU Yue, et al. An adaptive threshold algorithm for detectingulva prokifera in Sounthern Yellow Sea by remote sensing[J].Remote Sensing Information,2021,36(2):120-129.
9 王怡人,王胜强,喻樾 等.一种提取南黄海浒苔的自适应阈值遥感算法[J].遥感信息,2021,36(2):120-129.
10 SCANLAN C M, FODEN J, WELLS E, et al. The monitoring of opportunistic macroalgal blooms for the water framework directive[J]. Marine pollutionbulletin,2007,55(1-6):162-167. DOI: 10.1016/j.marpolbul.2006.09.017
doi: 10.1016/j.marpolbul.2006.09.017
11 WU Shihong. Research progress of remote sensing monitoring key technologies for urban black and odorous water bodies[J].Chinese Journal of Environmental Engineering,2019,13(6):1261-1271.
11 吴世红.城市黑臭水体遥感监测关键技术研究进展[J].环境工程学报,2019,13(6):1261-1271.
12 CAO Yun, HANG Xin, GAO Yi, et al. Remote sensing monitoring of urban black and odorous water bodies using GF-2 images:Taking the main urban area of Nanjing as an example[J]. Sichuan Environment, 2023,42(1):208-217.
12 曹云,杭鑫,高艺 等.利用高分二号影像对城市黑臭水体遥感监测--以南京市主城区为例[J].四川环境,2023,42(1):208-217.
13 LU Yingcheng, LIU Jianqiang, DING Jing, et al. Optical remote identification of spilled oils from the SANCHI oil tanker collision in the East China Sea[J]. Chinese Science Bulletin, 2019,64(31):3213-3222.
13 陆应诚,刘建强,丁静 等.中国东海“桑吉”轮溢油污染类型的光学遥感识别[J].科学通报,2019,64(31):3213-3222.
14 REN Guangbo, GUO Jie, MA Yi, et al. Oil spill detection and slick thickness measurement Via UAV hyperspectral imaging[J]. Acta Oceanologica Sinica, 2019,41(5):146-158.
14 任广波,过杰,马毅,等.海面溢油无人机高光谱遥感检测与厚度估算方法[J].海洋学报,2019,41(5):146-158.
15 FINGAS M, BROWN C. Review of oil spill remote sensing[J]. Marine Pollution Bulletin,2014,83(1):9-23. DOI:10. 1016/S1353-2561(98)00023-1
doi: 10. 1016/S1353-2561(98)00023-1
16 SU H, WU Z, ZHANG H, et al. Hyperspectral anomaly detection: A survey[J]. IEEE Geoscience and Remote Sensing Magazine,2021,10(1):64-90.DOI: 10.1109/MGRS. 2021. 3105440
doi: 10.1109/MGRS. 2021. 3105440
17 TAO X, ZHENG Y, CHEN W, et al. SVDD-based weighted oversampling technique for imbalanced and overlapped dataset learning[J]. Information Sciences: An International Journal, 2022(588):13-51.
18 WEI S A. Flexible region of interest extraction algorithm with adaptive threshold for 3-D synthetic aperture radar images[J]. Remote Sensing, 2021, 13(21):4308. DOI:10.3390/rs 13214308
doi: 10.3390/rs 13214308
19 PENG M. Adaptive subspace signal detection in structured interference plus compound Gaussian Sea clutter[J]. Remote Sensing, 2022, 14(9):2274. DOI:10.3390/rs14092274
doi: 10.3390/rs14092274
20 YANG Yang, LI Tiekun, YANG Shuwen, et al. Change detection of GF-1 remote sensing image based on spatial fuzzy C-means clustering and Bayesian Network[J]. Geomatics & Spatial Information Technology, 2023, 46(4):34-37,42
20 杨洋,李轶鲲,杨树文 等.基于空间模糊C均值聚类和贝叶斯网络的高分一号遥感影像变化检测[J].测绘与空间地理信息,2023,46(4):34-37,42.
21 WANG Qiao. Research framework of remote sensing monitoring and real-time diagnosis of earth surface anomalies[J]. Acta Geodaetica et Cartographica Sinica, 2022,51(7):1141-1152.
21 王桥.地表异常遥感探测与即时诊断方法研究框架[J].测绘学报,2022,51(7):1141-1152.
22 LIU F T, TING K M, ZHOU Z H. Isolation-based anomaly detection[J]. ACM Transactions on Knowledge Discovery from Data,2012,6(1):1-39. DOI:10.1145/2133360.2133363
doi: 10.1145/2133360.2133363
23 SONG Xiangyu. Research on anomaly detection in hyperspectral remote sensing images by Isolation Forest[D]. ChangChung:Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences,2022.
23 宋向宇. 基于孤立森林算法的高光谱遥感图像异常目标检测方法研究[D].长春:中国科学院大学(中国科学院长春光学精密机械与物理研究所),2022.
24 XUE Yuanyuan, HUANG Yuancheng, SU Yuanchao. Hyperspectral anomaly detection based on isolation forest with spatial weighting[J]. Science of Surveying and Mapping,2021,46(7):92-98.
24 薛园园,黄远程,苏远超.空间加权的孤立森林高光谱影像异常目标检测[J].测绘科学,2021,46(7):92-98.
25 LIU Lu, LUO Nianxue, ZHAO Qiansheng. Prediction of the outbreak scale of enteromorpha prolifera in the Yellow Sea based on historical data[J]. Surveying and Mapping Bulletin,2022(7):7-11.
25 刘璐,罗年学,赵前胜.基于历史数据的黄海浒苔爆发规模预测[J].测绘通报,2022(7):7-11.
26 CHEN Shuai. Remote sensing recognition of black and odorous water bodies based on Landsat 8 Images-A case study in Changsha[D]. ChangSha: Changsha University of Science and Technology,2021.
26 陈帅. 基于Landsat 8影像的黑臭水体遥感识别[D].长沙:长沙理工大学,2021.
27 LIAO Guoxiang. Numerical simulation of the transport and diffusion of spilled oil released from ‘Deepwater Horizon’ accident in the gulfof mexico[J]. Ocean Development and Management,2022,39(4):89-96.
27 廖国祥.“深水地平线”事故深海溢油输移扩散的数值模拟[J].海洋开发与管理,2022,39(4):89-96.
28 The Ministry of Housing and Urban-Rural Development of the People's Republic of China. Notice of the General Office of the Ministry of Housing and Urban-Rural Development and the General Office of the Ministry of Environmental Protection on Announcing the Investigation of Black and Odorous Water Bodies in Cities Across the Country[EB/OL] ,2016,2022.中华人民共和国住房和城乡建设部.住房城乡建设部办公厅 环境保护部办公厅关于公布全国城市黑臭水体排查情况的通知[EB/OL] ,2016,2022.
29 JIA Shaolin. The United states oil foreign policy to latin america during World War II[D]. Kaifeng: Henan University,2020.
29 贾少林. 二战时期美国对拉丁美洲的石油外交政策[D].开封:河南大学,2020.
30 WARD C H, TUNNELL J W. Habitats and biota of the Gulf of Mexico: An overview[R]. Habitats and Biota of the Gulf of Mexico: Before the Deepwater Horizon Oil Spill: Volume 1: Water Quality, Sediments, Sediment Contaminants, Oil and Gas Seeps, Coastal Habitats, Offshore Plankton and Benthos, and Shellfish, 2017: 1-54.
31 ZHAO Dong.multi-spectral remote sensing technologies slicks based on hyperspectral and researches on identifying sea surface oil[D].Wuhan: China University of Geosciences,2019.
31 赵冬. 基于高/多光谱遥感技术的海表油膜识别方法研究[D].武汉:中国地质大学,2019.
32 LU Y, LI X, TIAN Q,et al. Progress in marine oil spill optical remote sensing: Detected targets, spectral response characteris-tics,and theories[J]. Marine Geodesy,2013,36(3):334-346.
33 LIU F T, TING K M, ZHOU Z H. Isolation Forest[C]∥ Proceedings of the 2008 IEEE International Conference on Data Mining. IEEE Computer Society. 2008.
34 CAO Y. A method based on improved IForest for trunk extraction and denoising of individual street trees[J]. Remote Sensing, 2022, 15(1): 115.DOI:10.3390/rs15010115
doi: 10.3390/rs15010115
[1] 任慧敏,宋冬梅,王斌,甄宗晋,刘斌,张婷. 基于新极化特征参数的SAR海洋溢油检测[J]. 遥感技术与应用, 2020, 35(4): 934-942.
[2] 七珂珂,申茜,罗小军,李家国,姚月,杨崇. 基于GF-2影像的沈阳市黑臭水体遥感分级识别[J]. 遥感技术与应用, 2020, 35(2): 424-434.
[3] 孙晓,吴孟泉,何福红,张安定,赵德恒,李勃 . 2015年黄海海域浒苔时空分布及台风“灿鸿”影响研究[J]. 遥感技术与应用, 2017, 32(5): 921-930.
[4] 孙景,唐丹玲,潘刚. Radarsat2全极化数据在监测南海溢油事件中的方法研究[J]. 遥感技术与应用, 2016, 31(6): 1181-1189.
[5] 王晶,王丽姣,崔建涛,厉小润. 基于信噪比估计的波段选择与高光谱异常检测[J]. 遥感技术与应用, 2015, 30(2): 292-297.
[6] 王少波,严卫,艾未华,王蕊,刘天宁. 基于微波辐射计的海面溢油监测研究[J]. 遥感技术与应用, 2014, 29(1): 94-99.
[7] 宋莎莎,赵宇鹏,苏腾飞,马佑军,安 伟,孟俊敏,赵朝方. 海上溢油SAR卫星遥感监测系统研发[J]. 遥感技术与应用, 2013, 28(5): 928-933.
[8] 魏铼,胡卓玮. SAR影像海洋表面溢油检测方法与实现[J]. 遥感技术与应用, 2013, 28(1): 34-43.
[9] 焦俊超,马安青,娄安刚,陈爽,杨少磊. GIS和Google Earth开发在溢油预测中的整合应用[J]. 遥感技术与应用, 2011, 26(2): 215-219.
[10] 杨 红, 唐泽艳. 基于NOAA18的渤海烟台海域溢油遥感研究[J]. 遥感技术与应用, 2010, 25(2): 216-220.
[11] 邹亚荣,王华,邹斌 . 基于CFAR海上溢油检测研究[J]. 遥感技术与应用, 2008, 23(6): 629-632.
[12] 熊文成,吴传庆,魏斌,申文明,孙中平. SAR图像在韩国溢油监测中的应用[J]. 遥感技术与应用, 2008, 23(4): 413-413.
[13] 李智勇,郁文贤,匡纲要,吴 昊. 基于高维几何特性的高光谱异常检测算法研究[J]. 遥感技术与应用, 2003, 18(6): 379-383.
[14] 赵冬至,丛丕福. 海面溢油的可见光波段地物光谱特征研究[J]. 遥感技术与应用, 2000, 15(3): 160-164.