基于多尺度特征融合网络的新疆积雪覆盖度估算

doi:10.19678/j.issn.1000-3428.0061090

摘要/Abstract

摘要： 受复杂地形和遥感数据低分辨率的影响，传统的二值化积雪遥感产品在山区和林区的积雪覆盖度计算中存在严重误算和漏算的问题，从而导致积雪覆盖度估算精度低。基于风云四号A星多通道辐射扫描计（AGRI）新疆地区的遥感影像数据，提出一种多尺度特征融合网络的积雪覆盖度估算方法。通过深度残差网络和特征金字塔模式对卷积层各个阶段的特征信息进行重构，融合深层和浅层特征的多重语义信息，同时结合AGRI数据高时间分辨率的特性，拟合光谱信息和地理因素间的非线性关系，从而提高数据源和特征信息的整体利用率。实验结果表明，相比MOD10_FSC、BP-ANN_FSC和ResNet_FSC方法，该方法在A1~A4样本区中相关系数均值和解释回归模型的方差得分均值最高可提高8和6个百分点，且其均方误差均值仅为0.1，能够获得较高精度的积雪覆盖度估算结果。

关键词: 遥感数据, 风云四号静止卫星, 积雪覆盖度, 深度学习, 中分辨率成像仪

Abstract: Binary fractional snow remote sensing products are affected by complex terrain and low resolution of remote sensing data.The traditional binary fractional snow remote sensing products have serious miscalculation and omission in the calculation of Fractional Snow Cover(FSC) in mountainous and forest areas, resulting in low accuracy of FSC estimation.Based on the remote sensing image data of the Advanced Geosynchronous Radiation Imager(AGRI) of FY-4A in Xinjiang, a FSC estimation method based on multi-scale feature fusion network is proposed.The feature information of each stage of convolution layer is reconstructed through depth residual network and feature pyramid mode, and the multiple pieces of semantic information from deep and shallow feature layers are fused.Combined with the characteristics of high time resolution of AGRI data, the proposed method can better fit the nonlinear relationship between spectral information and geographical factors, improving the overall utilization of data source and feature information.The experimental results show that compared with the MOD10_FSC, BP-ANN_FSC and ResNet_FSC methods, in the four sample areas A1~A4, the mean value of correlation coefficient and the mean value of the variance score of the interpretation regression model are increased by up to 8 and 6 percentage points, and the mean square error is only 0.1, providing more accurate FSC estimation results.

Key words: remote sensing data, FY-4 geostationary satellite, Fractional Snow Cover(FSC), Deep Learning(DL), Moderate-resolution Imaging Spectroradiometer(MODIS)

中图分类号:

TP79

张永宏, 许帆, 阚希, 曹海啸. 基于多尺度特征融合网络的新疆积雪覆盖度估算[J]. 计算机工程, 2022, 48(3): 288-295.

ZHANG Yonghong, XU Fan, KAN Xi, CAO Haixiao. Xinjiang Fractional Snow Cover Estimation Based on Multi-Scale Feature Fusion Network[J]. Computer Engineering, 2022, 48(3): 288-295.

http://www.ecice06.com/CN/Y2022/V48/I3/288

图/表 15

20220401123834

20220401123837

20220401123841

20220401123844

20220401123847

20220401123851

20220401123854

20220401123858

20220401123902

20220401123905

20220401123909

20220401123912

20220401123916

20220401123919

20220401123923

参考文献

[1] ANDREWS J T, EVANS L W, WILLIAMS K M, et al.Cryosphere/ocean interactions at the margin of the laurentide ice sheet during the younger dryas chron:SE baffin shelf, northwest territories[J].Paleoceanography, 1991, 6(2):291.
[2] 王慧, 王胜利, 余行杰, 等.1961-2017年基于地面观测的新疆积雪时空变化研究[J].冰川冻土, 2020, 42(1):72-80. WANG H, WANG S L, YU X J, et al.Spatiotemporal variation of snow cover in Xinjiang based on ground observation from 1961 to 2017[J].Journal of Glaciology and Geocryology, 2020, 42(1):72-80.(in Chinese)
[3] SALOMONSON V V, APPEL I.Estimating fractional snow cover from MODIS using the using the normalized difference snow index[J].Remote Sensing of Environment, 2004, 89(3):351-360.
[4] ROSENTHAL W, DOZIER J.Automated mapping of montane snow cover at subpixel resolution from the landsat thematic mapper[J].Water Resources Research, 1996, 32(1):115-130.
[5] PAINTER T H, RITTGER K, MCKENZIE C, et al.Retrieval of subpixel snow covered area gram size, and albedo from MODIS[J].Remote Sensing of Environment, 2009, 113(4):868-879.
[6] 施建成.MODIS亚像元积雪覆盖反演算法研究——纪念杰出的地理学家、冰川学家施雅风先生逝世一周年[J]. 第四纪研究, 2012, 12(1):6-15. SHI J C.An automatic algorithm on estimating sub-pixel snow cover from modis[J].Quaternary Sciences, 2012, 12(1):6-15.(in Chinese)
[7] ZHAO H Y, HAO X H, ZHENG Z J, et al.Snow area proportion extraction algorithm based on FY-3d/MERSI- Ⅱ[J].Remote Sensing Technology and Application, 2018, 33(6):1004-1016.
[8] DOBREVA D I, KLEIN A G.Fractional snow cover mapping through artificial neural network analysis of MODIS surface reflectance[J].Remote Sensing of Environment, 2011, 115(12):3355-3366.
[9] MOOSAVI V, MALEKINEZHAD H, SHIRMOHAMMADI B.Fractional snow cover mapping from MODIS data using wavelet-artificial intelligence hybrid models[J].Journal of Hydrology, 2014, 511(7):160-170.
[10] HOU J L, HUANG C L.An application of ANN for mountainous snow cover fraction mapping with MODIS and ancillary topographic data[C]//Proceedings of IEEE International Geoscience and Remote Sensing Symposium.Washington D.C., USA:IEEE Press, 2014:1-10.
[11] 梁慧.青藏高原MODIS积雪面积比例制图算法研究[D].兰州:兰州大学, 2019. LIANG H.Research on MODIS snow area proportion mapping algorithm of Qinghai tibet plateau[D].Lanzhou:Lanzhou University, 2019.(in Chinese)
[12] 王慧, 王梅霞, 王胜利, 等.1961-2017年新疆积雪期时空变化特征及其与气象因子的关系[J].冰川冻土, 2021, 43(1):61-69. WANG H, WANG M X, WANG S L, et al.Temporal and spatial variation characteristics of snow cover period in Xinjiang from 1961 to 2017 and its relationship with meteorological factors[J].Journal of Glaciology and Geocryology, 2021, 43(1):61-69.(in Chinese)
[13] 李吉玫, 张毓涛, 张云云.天山北坡融雪期林地、草地、裸地积雪特性及其影响因素分析[J].灌溉排水学报, 2021, 40(1):106-114. LI J M, ZHANG Y T, ZHANG Y Y.Characteristics of snow cover and its influencing factors in woodland, grassland and bare land during snowmelt period on the northern slope of Tianshan mountains[J].Journal of Irrigation and Drainage, 2021, 40(1):106-114.(in Chinese)
[14] 郭强, 韩琦, 冯小虎.新一代风云四号气象卫星任务规划研究与应用[J].电子测量技术, 2020, 43(23):40-45. GUO Q, HAN Q, FENG X H.Research and application of new generation FY-4 meteorological satellite mission planning[J].Electronic Measurement Technology, 2020, 43(23):40-45.(in Chinese)
[15] 刘定生, 陈元伟, 李景山.遥感卫星地面预处理系统技术发展模式探讨[J].遥感信息, 2008(5):87-91. LIU D S, CHEN Y W, LI J S.Discussion on technology development mode of remote sensing satellite ground preprocessing system[J].Remote Sensing Information, 2008(5):87-91.(in Chinese)
[16] 史明阳, 王鹏, 汪卫.有监督时间序列分割与状态识别算法[J].计算机工程, 2020, 46(5):131-138. SHI M Y, WANG P, WANG W.Supervised time series segmentation and state recognition algorithm[J].Computer Engineering, 2020, 46(5):131-138.(in Chinese)
[17] 周志飞, 吴金龙, 李轶昳, 等.基于多任务级联残差网络的枪支图像识别系统[J].计算机工程, 2022, 48(1):214-219. ZHOU Z F, WU J L, LI Y Y, et al.Gun image recognition system based on multi task cascade residual network[J].Computer Engineering, 2022, 48(1):214-219.(in Chinese)
[18] 陶施帆, 李玉峰, 黄煜峰, 等.基于深度残差网络和注意力机制的人脸检测算法[J].计算机工程, 2021, 47(11):276-282. TAO S F, LI Y F, HUANG Y F, et al.Face detection algorithm based on deep residual network and attention mechanism[J].Computer Engineering, 2021, 47(11):276-282.(in Chinese)
[19] 苏炯铭, 刘鸿福, 项凤涛, 等.深度神经网络解释方法综述[J].计算机工程, 2020, 46(9):1-15. SU J M, LIU H F, XIANG F T, et al.Review of deep neural network interpretation methods[J].Computer Engineering, 2020, 46(9):1-15.(in Chinese)
[20] 叶萌, 杨娟, 汪荣贵, 等.基于特征聚合网络的小样本学习方法[J].计算机工程, 2021, 47(3):77-82. YE M, YANG J, WANG R G, et al.Small sample learning method based on feature aggregation network[J].Computer Engineering, 2021, 47(3):77-82.(in Chinese)

选择文件类型/文献管理软件名称

选择包含的内容