首页 > 论文 > 光学学报 > 37卷 > 11期(pp:1128001--1)

基于深度残差网络的高光谱遥感数据霾监测

Hyperspectral Data Haze Monitoring Based on Deep Residual Network

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

霾监测是环境治理中的关键技术之一。目前地面观测站进行霾监测的耗费较大,基于多光谱遥感的霾识别精度较低。将深度学习用于高光谱遥感数据的霾监测,提出一种基于深度残差网络的高光谱霾监测方法,利用深度网络提取霾光谱曲线特征,再使用残差学习等方法降低网络训练难度,得到了霾监测模型。苏州地区Hyperion高光谱数据集上的实验表明,与其他遥感霾监测方法相比,所提方法的霾识别精度更高。

Abstract

Haze monitoring is one of the key technologies for environmental governance. At present, the cost of the ground haze monitoring is very high and the accuracy of the multispectral remote sensing haze monitoring is low. The hyperspectral sensing data haze monitoring is studied by deep learning. A hyperspectral haze monitoring algorithm based on deep residual network is presented. The features of haze hyperspectral curves are obtained with the deep network. The difficulty of the network training is decreased with the residual leaning method, and a haze monitoring model is achieved. The experimental results of the Suzhou Hyperion hyperspectral data sets show that, compared with other methods of remote haze monitoring, the proposed method has higher recognition accuracy in haze monitoring.

投稿润色
补充资料

中图分类号:TP75

DOI:10.3788/aos201737.1128001

所属栏目:遥感与传感器

基金项目:国家自然科学基金(U1406404),上海市军民结合项目(沪经信军[2014年]495号)

收稿日期:2017-03-16

修改稿日期:2017-05-22

网络出版日期:--

作者单位    点击查看

陆永帅:上海交通大学航空航天学院, 上海 200240
李元祥:上海交通大学航空航天学院, 上海 200240
刘 波:上海卫星工程研究所十五室, 上海 201108
刘 辉:上海卫星工程研究所十五室, 上海 201108
崔林丽:上海市气象科学研究所卫星遥感应用技术研究室, 上海 200030

联系人作者:李元祥(yuanxli@sjtu.edu.cn)

备注:陆永帅(1991-),男,硕士研究生, 主要从事图像处理与机器学习的理论和应用方面的研究。

【1】Wang Zhongting, Chen Liangfu, Li Qing, et al. Simulation of multi-angle polarized reflectance of haze[J]. Acta Optica Sinica, 2015, 35(9): 0901002.
王中挺, 陈良富, 厉青, 等. 灰霾组分的多角度偏振反射特性模拟[J]. 光学学报, 2015, 35(9): 0901002.

【2】Liu Jianguo, Gui Huaqiao, Xie Pinhua, et al. Recent progress of atmospheric haze monitoring technology[J]. Journal of Atmospheric and Environmental Optics, 2015, 10(2): 93-101.
刘建国, 桂华侨, 谢品华, 等. 大气灰霾监测技术研究进展[J]. 大气与环境光学学报, 2015, 10(2): 93-101.

【3】Lee K H, Kim Y J, Kim M J. Characteristics of aerosol observed during two severe haze events over Korea in June and October 2004[J]. Atmospheric Environment, 2006, 40(27): 5146-5155.

【4】Ghauri B. Estimating area covered by Haze and fog in Pakistan and India during winters[C]. IEEE Conference on Geoscience and Remote Sensing Symposium (IGARSS), 2016: 16445492.

【5】Dai Yangyang, Li Chengfan, Zhou Shiqiang, et al. Haze monitoring of Shanghai area based on remote sensing[J]. Engineering of Surveying and Mapping, 2015, (12): 29-32.
戴羊羊, 李成范, 周时强, 等. 基于遥感的上海地区雾霾监测研究[J]. 测绘工程, 2015, (12): 29-32.

【6】Wang Zhongting, Li Qing, Li Shenshen, et al. The monitoring of haze from HJ-1[J]. Spectroscopy and Spectral Analysis, 2012, 32(3): 775-780.
王中挺, 厉青, 李莘莘, 等. 基于环境一号卫星的霾监测应用[J]. 光谱学与光谱分析, 2012, 32(3): 775-780.

【7】Liu Yonghong. Research on haze identification in Beijing based on NOAA/AVHRR satellite data[J]. Meteorological Monthly, 2014, 40(5): 619-627.
刘勇洪. 基于 NOAA/AVHRR 卫星资料的北京地区霾识别研究[J]. 气象, 2014, 40(5): 619-627.

【8】Niu Zhichun, Jiang Sheng, Li Xuwen, et al. The remote sensing monitoring operational system of haze pollution in Jiangsu province[J]. Environmental Monitoring & Forewarning, 2014, 6(5): 15-18.
牛志春, 姜晟, 李旭文, 等. 江苏省霾污染遥感监测业务化运行研究[J]. 环境监控与预警, 2014, 6(5): 15-18.

【9】Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[C]. Neural Information Processing Systems Conference, 2012: 1097-1105.

【10】Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]. Proceedings of the IEEE conference on computer vision and pattern recognition, 2014: 580-587.

【11】Liu Dawei, Han Ling, Han Xiaoyong. High spatial resolution remote sensing image classification based on deep learning[J]. Acta Optica Sinica, 2016, 36(4): 0428001.
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, 36(4): 0428001.

【12】Yin Baocai, Wang Wentong, Wang Lichun. Review of deep learning[J]. Journal of Beijing University of Technology, 2015, 1(1): 48-59.
尹宝才, 王文通, 王立春. 深度学习研究综述[J]. 北京工业大学学报, 2015, 1(1): 48-59.

【13】He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016: 770-778.

【14】Xu Wenbin, Zheng Xiaobing, Yi Weining. Cross-calibration method based on hyperspectral imager hyperion[J]. Acta Optica Sinica, 2013, 33(5): 0528002.
徐文斌, 郑小兵, 易维宁. 基于超光谱成像仪Hyperion的交叉定标方法[J]. 光学学报, 2013, 33(5): 0528002.

【15】Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[C]. Proceedings of the 32nd International Conference on Machine Learning, 2015, 37: 448-456.

【16】Hinton G E, Srivastava N, Krizhevsky A, et al. Improving neural networks by preventing co-adaptation of feature detectors[J]. Computer Science, 2012, 3(4): 212-223.

【17】Glorot X, Bordes A, Bengio Y. Deep sparse rectifier neural networks[C]. Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, 2011, 15: 315-323.

【18】Bottou L. Large-scale machine learning with stochastic gradient descent[C]. 19th International Conference on Computational Statistics, 2010: 177-186.

引用该论文

Lu Yongshuai,Li Yuanxiang,Liu Bo,Liu Hui,Cui Linli. Hyperspectral Data Haze Monitoring Based on Deep Residual Network[J]. Acta Optica Sinica, 2017, 37(11): 1128001

陆永帅,李元祥,刘 波,刘 辉,崔林丽. 基于深度残差网络的高光谱遥感数据霾监测[J]. 光学学报, 2017, 37(11): 1128001

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF