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

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.

[1]  Thorsten Joachims,et al.  Text Categorization with Support Vector Machines: Learning with Many Relevant Features , 1998, ECML.

[2]  Vladimir Vapnik,et al.  Methods of Pattern Recognition , 2000 .

[3]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[4]  Daphne Koller,et al.  Support Vector Machine Active Learning with Applications to Text Classification , 2000, J. Mach. Learn. Res..

[5]  Lorenzo Bruzzone,et al.  Classification of hyperspectral remote sensing images with support vector machines , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[6]  S. Ameur,et al.  Cloud classification using the textural features of Meteosat images , 2004 .

[7]  G. Stephens.  Cloud Feedbacks in the Climate System: A Critical Review , 2005 .

[8]  Evgueni I. Kassianov,et al.  Cloud Sky Cover versus Cloud Fraction: Whole-Sky Simulations and Observations , 2005 .

[9]  Maneesha Singh,et al.  Automated ground-based cloud recognition , 2005, Pattern Analysis and Applications.

[10]  Cordelia Schmid,et al.  Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive Study , 2006, 2006 Conference on Computer Vision and Pattern Recognition Workshop (CVPRW'06).

[11]  Josep Calbó,et al.  Retrieving Cloud Characteristics from Ground-Based Daytime Color All-Sky Images , 2006 .

[12]  Filomena Romano,et al.  Physical and statistical approaches for cloud identification using Meteosat Second Generation-Spinning Enhanced Visible and Infrared Imager Data , 2008 .

[13]  Josep Calbó,et al.  Feature Extraction from Whole-Sky Ground-Based Images for Cloud-Type Recognition , 2008 .

[14]  E. Pereira,et al.  The Use of Euclidean Geometric Distance on RGB Color Space for the Classification of Sky and Cloud Patterns , 2010 .

[15]  R. Suganya,et al.  Data Mining Concepts and Techniques , 2010 .

[16]  A. Heinle,et al.  Automatic cloud classification of whole sky images , 2010 .

[17]  Lei Liu,et al.  Cloud Classification Based on Structure Features of Infrared Images , 2011 .

[18]  Ahmed A. K. Tahir.  A System Based on Ratio Images and Quick Probabilistic Neural Network for Continuous Cloud Classification , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[19]  Jun Yang,et al.  A Hybrid Thresholding Algorithm for Cloud Detection on Ground-Based Color Images , 2011 .

[20]  James Zijun Wang,et al.  Thin Cloud Detection of All-Sky Images Using Markov Random Fields , 2012, IEEE Geoscience and Remote Sensing Letters.

[21]  Weitao Lu,et al.  An Automated Cirrus Cloud Detection Method for a Ground-Based Cloud Image , 2012 .

[22]  George Economou,et al.  Cloud detection and classification with the use of whole-sky ground-based images , 2012 .

[23]  Alexandros G. Charalambides,et al.  Equipment and methodologies for cloud detection and classification: A review , 2013 .

[24]  Yunxue Shao,et al.  Salient local binary pattern for ground-based cloud classification , 2013, Acta Meteorologica Sinica.

[25]  J. Kleissl,et al.  Development of a sky imaging system for short-term solar power forecasting , 2014 .

[26]  Hsu-Yung Cheng,et al.  Block-based cloud classification with statistical features and distribution of local texture features , 2014 .

[27]  Development of a sky imaging system , 2014 .

[28]  Tieniu Tan,et al.  Feature Coding in Image Classification: A Comprehensive Study , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[29]  Jeffrey J. Rodriguez,et al.  A New Contrast-Enhancing Feature for Cloud Detection in Ground-Based Sky Images , 2015 .

[30]  Shuang Liu,et al.  Learning Discriminative Salient LBP for Cloud Classification in Wireless Sensor Networks , 2015, Int. J. Distributed Sens. Networks.

[31]  Apichat Heednacram,et al.  Feature extraction techniques for ground-based cloud type classification , 2015, Expert Syst. Appl..

[32]  Wen Yao,et al.  An automated cloud detection method based on the green channel of total-sky visible images , 2015 .

[33]  Yan Wang,et al.  Automatic Recognition of Cloud Images by Using Visual Saliency Features , 2015, IEEE Geoscience and Remote Sensing Letters.

[34]  Ying Ma,et al.  A hybrid method based on extreme learning machine and k-nearest neighbor for cloud classification of ground-based visible cloud image , 2015, Neurocomputing.