基于机器学习的重庆长江航道雾图像特征识别研究

doi:10.3969/j.issn.1673-503X.2021.01.014

Abstract

Abstract:

Based on the image data of fog events taken by radar stations on the Yangtze River waterways in Chongqing, images without fog events and with five types of fog events were trained using algorithms including K-nearest neighbor, support vector machine, back propagation neural network, and random forest.According to the training results, a fog image identification model was built, and the identification accuracy was tested.The results show that machine learning can effectively identify fog images, and random forest performances better than the other three algorithms.The model has a recognition accuracy of 100% for non-fog recognition, over 90% for light fog and strong dense fog events, and over 70% for fog, heavy fog, and dense fog events.

Key words: Fog, Machine learning, Pattern recognition, Graphical user interface

CLC Number:

P409

Yuan-mou WANG,Jia-qi LI,Shi-ji CHEN,Jia-ping TANG,Bai-cheng XIA,Shi-gang HAN. Identification of the fog image features on the Yangtze River waterways in Chongqing based on machine learning[J]. Journal of Meteorology and Environment, 2021, 37(1): 106-112.

Figures/Tables 9

Fig.1

Fig.2

Fig.3

Table 1

Table 2

Fig.4

Table 3

Fig.5

Fig.6

References 21

1	赵慧霞, 王维国, 李泽椿, 等. 雾对我国交通运输的不利影响及对策[J]. 气象与环境学报, 2010, 26 (2): 58- 62.
2	居为民, 孙涵. 卫星遥感资料在沪宁高速公路大雾监测中的初步应用[J]. 遥感信息, 1997, (3): 25- 27.
3	李亚春, 孙涵, 李湘阁, 等. 用GMS-5气象卫星资料遥感监测白天雾的研究[J]. 南京气象学院学报, 2001, 24 (3): 343- 349.
4	梁超圣, 柳健, 王志成. 低能见度前下视航拍图像中机场跑道的自动检测[J]. 华中科技大学学报: 自然科学版, 2005, 33 (3): 27- 29.
5	孙晓宁, 陆文骏. 一种新的雾天图像显著性检测方法[J]. 计算机与数字工程, 2015, 43 (11): 2029- 2034.
6	李才媛, 韦惠红, 邓红. SVM方法在武汉市大雾预警预报中的应用[J]. 暴雨灾害, 2008, 27 (3): 264- 267.
7	项文书. 基于交通视频的能见度估计研究[D]. 上海: 上海交通大学, 2014.
8	史达伟, 李超, 史逸民, 等. 基于机器学习的大雾天气背景下特强浓雾本地化诊断研究[J]. 灾害学, 2018, 33 (2): 193- 199.
9	马啸, 邵利民, 徐冠雷, 等. 基于K均值聚类算法的雾霾天气自动识别[J]. 舰船电子工程, 2018, 38 (12): 124- 128.
10	周建平, 苗开超, 江春, 等. 高速公路夜间大雾图像特征及其识别[J]. 气象科技, 2019, 47 (1): 29- 34.
11	朱熙文. 大数据下的在线机器学习算法研究与应用[D]. 成都: 西南交通大学, 2017.
12	闫蕾芳. 基于深度学习的图像分类的研究[D]. 济南: 山东大学, 2017.
13	骞宇澄, 刘昭策. 深度学习的实现与发展-从神经网络到机器学习[J]. 电子技术与软件工程, 2017, (11): 30- 31.
14	林丹. 基于小波变换的K-means算法在遥感图像分类中的应用研究[D]. 邯郸: 河北工程大学, 2015.
15	居红云, 张俊本, 李朝峰, 等. 基于K-means与SVM结合的遥感图像全自动分类方法[J]. 计算机应用研究, 2007, 24 (11): 324- 326.
16	贺皓, 罗慧. 基于支持向量机模式识别的大雾预报方法[C]//第27届中国气象学会年会论文集, 北京: 中国气象学会, 2012: 1-6.
17	刘年庆, 蒋建莹, 吴晓京. 基于支持向量机的遥感大雾判识[J]. 气象, 2007, 33 (10): 75- 79.
18	陈贝, 徐洪刚, 王明天, 等. 成乐高速公路大雾预报方法研究[J]. 高原山地气象研究, 2012, 32 (2): 70- 76.
19	袁志康, 张韧, 周树道, 等. 基于人工神经网络算法的某机场雾的模式识别预报[J]. 解放军理工大学学报: 自然科学版, 2002, 3 (4): 82- 86.
20	金宝刚, 张韧, 王晓蕾, 等. 基于BP神经网络与纹理分析优化的雾检测[J]. 解放军理工大学学报: 自然科学版, 2009, 10 (2): 195- 199.
21	包红军, 陈辉, 田华, 等. 京珠高速公路低能见度(雾)神经网络预测模型[C]//第28届中国气象学会年会-S10公共气象服务政策体制机制和学科建设. 厦门: 中国气象学会, 2011.

天气类型	无雾	轻雾	雾	大雾	浓雾	强浓雾
饱和度方差	(500, +∞)	(300, 500]	(100, 300]	(100, 300]	(100, 300]	(0, 100]
熵	(0.26, +∞)	(0.26, +∞)	(0.25, 0.26]	(0.24, 0.25]	(0.21, 0.24]	(0, 0.21]
样本数	2532	158	284	291	216	209

天气类型	无雾	轻雾	雾	大雾	浓雾	强浓雾
能量	0.128302	0.10812	0.126735	0.128953	0.126922	0.141890
熵	0.299417	0.280122	0.261399	0.249963	0.225267	0.205715
对比度	2.461470	2.596750	2.411628	2.368502	2.341374	2.212856
相关性	0.947251	0.962863	0.963829	0.970463	0.974697	0.976354
饱和度均值	56.06066	18.5819	12.62698	12.03518	11.22345	9.212304
饱和度方差	1159.287	425.4326	191.4073	181.1989	151.9692	96.43077

天气类型	K最近邻	SVM	BP神经网络	随机森林(1000棵)	随机森林(5000棵)	随机森林(50000棵)
无雾	100	100	100	100	100	100
轻雾	100	95.83	79.17	100	100	100
雾	78.33	48.33	76.67	93.33	91.67	91.67
大雾	77.19	66.67	40.35	89.47	89.47	89.47
浓雾	72.22	51.85	75.93	81.48	5.93	77.78
强浓雾	89.74	84.62	0	97.44	97.44	97.44

[1]	Qi YAN, Shuang LI, Fang-da TENG, Fang-ni WU, Shu-e LIANG, Ai-zhong ZHANG. Diagnoses of two foggy processes in Liaoning province in 2018 [J]. Journal of Meteorology and Environment, 2020, 36(6): 91-97.
[2]	Cui-yan ZHANG,Qi YAN,Xue BAI,Guang-mei ZHANG,Fang-lei LOU,Fu-ran ZHOU,Shu-ting CHANG. Climate characteristics and causes of heavy fog in Jinzhou from 1999 to 2018 [J]. Journal of Meteorology and Environment, 2020, 36(3): 87-92.
[3]	Kai NIE, Jing-li WANG, Ming LEI, Wei HAN, Jin-cheng LIU. Comparison of visibility in rainfall and haze-fog weather in Beijing using digital photographic visibility system [J]. Journal of Meteorology and Environment, 2020, 36(1): 28-35.
[4]	Bo-ni WANG,Xue-rong ZHANG,Ming SUN,Li TIAN,Mei-juan PU. Characteristics and formation mechanism of precipitation fog events in Jiangsu province [J]. Journal of Meteorology and Environment, 2020, 36(1): 58-66.
[5]	Wan-ying ZHANG,Xiu-bo SUN,Lan LI,Jian-qiang ZHANG,Li SUN,Xin QI. Climate characteristics and meteorological impact factors of heavy fog along Liaoning coast expressway [J]. Journal of Meteorology and Environment, 2020, 36(1): 67-73.
[6]	WU Jie, ZHOU Jian-ping, DENG Xue-liang, WANG Chuan-hui, ZHOU Wen-lin. Analysis of visibility characteristics and its influencing factors over Chaohu Lake area [J]. Journal of Meteorology and Environment, 2019, 35(2): 40-45.
[7]	GUO Qu, LI Rui, RUAN Ding-lun, ZHANG Tian-yn. Climatic variation of winter fog and its causes in Chongqing area [J]. Journal of Meteorology and Environment, 2018, 34(5): 47-56.
[8]	CHEN Yue-hao, XIONG Ming-ming, CAO Jing-fu, YANG Yan-juan. Quantitative study on the effect of haze-fog events on solar radiation in Tianjin [J]. Journal of Meteorology and Environment, 2018, 34(5): 25-30.
[9]	CHEN Xiang-xiang, XU Ai-hua, XU Bin, XIAO An. Characteristics of spatiotemporal distribution and influence elements of three kinds of regional heavy fog in Jiangxi province from 2000 to 2012 [J]. Journal of Meteorology and Environment, 2018, 34(3): 37-47.
[10]	XIA Fan, LI Chang-yi. Fog weather forecast experiments over Shandong province based on three visibility schemes [J]. Journal of Meteorology and Environment, 2018, 34(3): 48-57.
[11]	ZHANG Chi, LIU Jun, LI Xu-guang, ZHANG Qi-hai, YANG Bi-xuan, YANG Jun. A cloud classification method based on information fusion of visible and infrared images [J]. Journal of Meteorology and Environment, 2018, 34(1): 82-90.
[12]	YU Zhi-ming, WANG Si-yao, MA Dong-liang. Study on weather classification and forecasting index of marine meteorological disasters in the Bohai sea [J]. Journal of Meteorology and Environment, 2018, 34(1): 106-111.
[13]	ZHU Jia-rong, QI Liang-bo. Evolution and forecast of a radiation fog event in Shanghai [J]. Journal of Meteorology and Environment, 2017, 33(1): 1-11.
[14]	JIN Wei, QU Shu-lin, WANG Ye-xiu, SHEN Yu-min, DAI Ping, GAO Ling-yun, QU Yan, YUAN Jiu-song. Climatic characteristics and causes of heavy fog in Anshan area [J]. Journal of Meteorology and Environment, 2016, 32(6): 69-76.
[15]	LI Yang, MA Shu-qing, JIA Xiao-fang, ZHENG Yu, LU Sai, ZHAO Hu-jia. Vertical distribution of atmospheric particulate matter during a heavy fog-haze event in Beijing observed by an unmanned aerial vehicle [J]. Journal of Meteorology and Environment, 2016, 32(6): 179-183.

Identification of the fog image features on the Yangtze River waterways in Chongqing based on machine learning

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 21

Related Articles 15

Metrics

Comments

Recommended 10