Open AccessANALYSIS OF CLOUDINESS BY SEGMENTATION AND MONITORING OF SATELLITE MAP IMAGES
Author(s) -
Roman Melnyk,
Yurii Kalychak,
Roman Kvit
Publication year - 2019
Publication title -
computing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.184
H-Index - 11
eISSN - 2312-5381
pISSN - 1727-6209
DOI - 10.47839/ijc.18.2.1415
Subject(s) - computer science , histogram , cloud cover , segmentation , cluster analysis , artificial intelligence , image segmentation , clipping (morphology) , computer vision , remote sensing , geography , cloud computing , image (mathematics) , linguistics , philosophy , operating system
The algorithm of the dynamic threshold segmentation of images using clipping plane in a three-dimensional XYZ image space is proposed. To build the clipping plane of the dynamic threshold the precession and nutation angles as the base threshold values are found. The developed algorithm is applied to the satellite map images to get cloudiness intensity. The satellite map images are transformed by segmentation and inversion. The segmented and inverted images are scanned to receive the distributed cumulative histograms. By the help of so-called cloudiness meter the statistical data is processed for calculation and monitoring of cloudiness in Ukraine. The formulas to create an image of the distributed cumulative histogram are considered. Formulas to reconstruct images of the rotated satellite map images are proposed. The satellite weather map images were taken from the Wunderground services. The clustering algorithm is used to classify the regions of Ukraine by cloudiness intensity, which were created distributed cumulative images. The clustering algorithm is based on the agglomerative procedure.