Content area
Full Text
ABSTRACT
Some cloud structure features that can be extracted from infrared images of the sky are suggested for cloud classification. Both the features and the classifier are developed over zenithal images taken by the whole-sky infrared cloud-measuring system (WSIRCMS), which is placed in Nanjing, China. Before feature extraction, the original infrared image was smoothed to suppress noise. Then, the image was enhanced using top-hat transformation and a high-pass filtering. Edges are detected from the enhanced image after adaptive optimization threshold segmentation and morphological edge detection. Several structural features are extracted from the segment image and edge image, such as cloud gray mean value (ME), cloud fraction (ECF), edge sharpness (ES), and cloud mass and gap distribution parameters, including very small-sized cloud mass and gaps (SMG), middle-sized cloud gaps (MG), medium-small-sized cloud gaps (MSG), and main cloud mass (MM). It is found that these features are useful for distinguishing cirriform, cumuliform, and waveform clouds. A simple but efficient supervised classifier called the rectangle method is used to do cloud classification. The performance of the classifier is assessed with an a priori classification carried out by visual inspection of 277 images. The index of agreement is 90.97%.
(ProQuest: ... denotes formulae omitted.)
1. Introduction
Cloud macroscopic parameters such as cloud cover and cloud type are traditionally observed by humans on the ground. In recent years, a number of ground-based sky imagers such as the whole-sky imager (WSI; Shields et al. 1998), total-sky imager (TSI; Long et al. 2001, 2006), infrared cloud imager (ICI; Shaw and Thurairajah 2003; Shaw et al. 2002; Thurairajah 2004), all-sky imager (ASI; Huo and Lu 2002; Cazorla et al. 2008), and whole-sky infrared cloud-measuring system (WSIRCMS; Sun et al. 2009a) have been developed to estimate cloud cover. Few studies focused on the cloud classification based on the ground-based sky images. Buch et al. (1995) suggested a method for classifying WSI images into the following five types: clear, stratus, cumulus, cirrus, and altocumulus. They suggested some texture features and obtained a misclassification rate of 39% when compared with visual classification. Peura et al. (1996) considered that the method for ground-based cloud classification should treat whole-sky images as compositions of objects. They used the visual appearance of clouds, such as sharpness of cloud edges,...