Preparatory Research on Automatic Cloud Type ClassificationTechnique Based on MODIS Data

PDF(4083 KB)
Plateau Meteorology ›› 2008, Vol. 27 ›› Issue (增刊) : 224-229.

Preparatory Research on Automatic Cloud Type ClassificationTechnique Based on MODIS Data

Author information +
History +

Abstract

It is the basic work to discriminatethe cloud types by use of remote sensing data in the several meteorological research fields. In this paper, the classification test betweencloud and backgroundas well as different cloud types was finished after the essentialpre\|processingand data analysis work were performed, at the same time the operational potentialof this method was discussed. The results show that the better classification effect can be got by use of traditional classificationmethod with texture features based on the gray co\|occurrencematrix theory, based on the statisticalanalysis the study show that the entropy feature among eight common used texturescan better describe the difference betweensea, land and differentcloud types especially for the snow which is frequently confusedwith the cloud type in the visible spectrum, at the same time, the research show that the entropy value show the very similar statisticcharacteristics in band1 and band4, so only one band betweenwhich can be selected in order to improve the processing speed.

Key words

MODIS / Texture features / Cloud classification

Cite this article

Download Citations
. Preparatory Research on Automatic Cloud Type ClassificationTechnique Based on MODIS Data. Plateau Meteorology. 2008, 27(增刊): 224-229

References

[1]Guillory A R, J M Lecue, G J Jedlovec, et al. Cloud filtering using a bi\|spectral spatial coherence approach[C]. 9th Conference on Satellite Meteorology and Oceanography, A M S, Paris, France, 1998: 374-376
[2]Gary J Jedlovec, Huntsville, Kevin Laws. Operational cloud detection in goes imagery[C]. 11th Conference on Satellite Meteorology and Oceanography, 2001: 357
[3]Derrien M. Automatic cloud detection applied to NOAA-11/AVHRR imagery [J]. Remote Sense of Environment, 1993, 46(3): 246-267
[4]Griffin M, H Burke, D Mandl, et al. Cloud cover detection algorithm for EO-1 hyperion imagery[C]. EO-1 SVT Meeting. Hilo: MIT Lincoln Laboratory NASA
[5]白慧卿, 方宗义, 吴蓉璋, 等.基于人工神经网络的GMS云图四类云系的识别应用[J].气象学报, 1998, 56(4):402-409
[6]杨澄, 袁招洪, 顾松山. 用多谱阈值法进行GMS-5卫星云图云型分类的研究[J]. 南京气象学院学报, 2002, 25(6): 747-754
[7]师春香, 吴蓉璋, 项续康. 多阈值和神经网络卫星云系自动分割试验[J]. 应用气象学报, 2001, 12(1): 70-77
[8]周伟, 李万彪.利用GMS-5红外资料进行云的分类识别[J]. 北京大学学报(自然科学版), 2003, 39(1): 83-90
[9]张韧, 王海俊, 孙照渤, 等. 双光谱卫星云图的模糊推理云分类[J].防灾减灾工程学报, 2004, 24(3): 257-263
[10]Baum. A grouped threshold approach for scene identification in AVHRR imagery[J].J Atmos Oceanic Technol, 1999, 16(6): 793-800
[11]谈建国, 周红妹, 陆贤, 等. NOAA卫星云图检测和云修复业务应用系统的研制和建立[J].遥感技术与应用, 2000, 15(4): 228-231
[12]MODIS Cloud Mask Product Description. http://modis\|atmos.gsfc.nasa.gov/MOD35_L2/index.html
[13]鄢俊洁, 刘良明. MODIS云检测研究[C]. 第十四届全国遥感技术学术交流会论文摘要集, 2003
[14]汤琦, 毛节泰, 李成才. 利用MODIS资料对中国西部地区的云检测[J]. 高原气象, 2006, 25(6): 990-1000
[15]盛夏, 孔龙祥, 郑庆梅. 利用MODIS数据进行云检测[J]. 解放军理工大学学报(自然科学版), 2004, 5(4): 99-102
[16]盛夏, 孙龙祥, 郑庆梅. 模拟退火优化BP神经网络进行云相态分类[J].解放军理工大学学报(自然科学版), 2008, 9(1): 98-102
[17]Haralick R M, K Shanmugan, I Dinstein. Textural features for image classification[J]. IEEE Transactions on Systems, Man and Cybernetics, 1973, 3(6): 610-621
[18]寿亦萱, 张颖超, 赵忠明. 暴雨过程的卫星云图纹理特征研究[J]. 南京气象学院学报, 2005, 28(3): 337-343
[19]蔡艳, 傅德胜. 基于卫星遥感图象纹理特征的云类识别方法及软件设计[J]. 南京气象学院学报, 1999, 22(3): 416-422
[20]王彦磊, 张韧, 孙照渤, 等. 基于模糊C均值聚类的云图样本修正与云类自动识别[J]. 海洋科学进展, 2005, 23(2): 219-226
PDF(4083 KB)

1316

Accesses

0

Citation

Detail

Sections
Recommended

/