From pixels to patches: a cloud classification method based on a bag of micro-structures

Qingyong Li,Jun Yang,Wen Yao,Weitao Lu,Ying Ma,Zhen Zhang

doi:10.5194/amt-9-753-2016

Abstract

Abstract. Automatic cloud classification has attracted more and more attention with the increasing development of whole sky imagers, but it is still in progress for ground-based cloud observation. This paper proposes a new cloud classification method, named bag of micro-structures (BoMS). This method treats an all-sky image as a collection of micro-structures mapped from image patches, rather than a collection of pixels. It represents the image with a weighted histogram of micro-structures. Based on this representation, BoMS recognizes the cloud class of the image by a support vector machine (SVM) classifier. Five classes of sky condition are identified: cirriform, cumuliform, stratiform, clear sky, and mixed cloudiness. BoMS is evaluated on a large data set, which contains 5000 all-sky images captured by a total-sky cloud imager located in Tibet (29.25° N, 88.88° E). BoMS achieves an accuracy of 90.9 % for 10-fold cross-validation, and it outperforms state-of-the-art methods with an increase of 19 %. Furthermore, influence of key parameters in BoMS is investigated to verify their robustness.

Highlights

Clouds play an important role in the hydrological cycle and the energy balance of the atmosphere–earth surface system because of the interaction with solar and terrestrial radiation (Stephens, 2005)
This study presents the new cloud classification method based on a bag of micro-structures, whereas most state-ofthe-art methods (Heinle et al, 2010; Liu and Zhang, 2015; Kliangsuwan and Heednacram, 2015; Cheng and Yu, 2015) apply traditional features based on pixels
An all-sky image is treated as a collection of micro-structures just as a document consists of words, and it is represented by a high-dimensional histogram of micro-structures

Summary

Introduction

Clouds play an important role in the hydrological cycle and the energy balance of the atmosphere–earth surface system because of the interaction with solar and terrestrial radiation (Stephens, 2005). Cloud classification is first investigated based on satellite images (Ameur et al, 2004; Tahir, 2011; Hu et al., 2015). Most of these methods apply texture features and classifier models to recognize cloud type. The information provided by large-scale satellite images is not sufficient enough. These images have too low resolution to capture detailed characteristics of local clouds; thin clouds and earth surface are frequently confused in satellite images because of their similar brightness and temperature (Ricciardelli et al, 2008). Ground-based cloud observation can obtain more accurate characteristics for local clouds, and ground-based cloud classification has attracted more and more attention (Tapakis and Charalambides, 2013)

Methods

Results

Conclusion