Cloud Classification Based on Structure Features of Infrared Images

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%.

Cloud detection in all-sky images via multi-scale neighborhood features and multiple supervised learning techniques

“…A category of research works are devoted to detecting (Long et al, 2006), classifying (Calbo and Sabburg, 2008;Heinle et al, 2010;Isosalo et al, 2007;Liu et al, 2011;Martínez-Chico et al, 2011;Zhuo et al, 2014), and tracking clouds (Marquez and Coimbra, 2013;Tapakis and Charalambides, 2013;Wood-Bradley et al, 2012). The relationships between cloud coverage and surface solar irradiance have been explored (Feister and Shields, 2005;Fu and Cheng, 2013;Pfister et al, 2003).…”

Fuzzy AutoEncode Based Cloud Detection for Remote Sensing Imagery

“…According to the unique characteristics of clouds, which are brighter and colder than most of the earth surface, spectral features can always be used for cloud pixel detection. In addition, some existing methods add other information of images, such as texture information, shape information, spatial information, and so on [25][26][27][28].…”

Unsupervised Multi-level Thresholding Method for Weather Satellite Cloud Segmentation

“…The evolution of image processing techniques -in recent decades -has achieved considerable advances in automatic processing of satellite imagery. Unsupervised segmentation and classification 1 of cloud structures from satellite images is one of the most studied applications in meteorology to improve weather forecasts [2,3,4,5,6,7]. The segmentation of cloud areas using simple thresholding techniques (use of a single threshold for the whole image) can not provide all the important structural information of the segmented areas.…”