Method of constructing a space-time grid of variable size images and its application in image processing of landfills

• Main Authors: Richter Andrey, Ignatiev Vladimir Y., Kazaryan Maretta, Voronin Viacheslav, Shakhramanyan Mihail
• Format: Article
• Language: English
• Published: Faculty of Technical Sciences in Cacak 2019-01-01
• Series: Serbian Journal of Electrical Engineering
• Subjects: satellite imagery; image slicing; sites; automation; time series; temporal processing; detection; index of the underlying surface; landfills; dumps; waste disposal facilities
• Online Access: http://www.doiserbia.nb.rs/img/doi/1451-4869/2019/1451-48691901123R.pdf
• ISSN: 1451-4869; 2217-7183

The paper proposes a method of slicing space images into areas of varying sizes to automate the image processing, beginning from the receipt of raw data (download pictures) up to mapping of the underlying surface indices. Indices calculation are based on one image (the surface and the vegetation indices temperature, indices of objects detection on the ground and etc.) and their time series (conducting temporal processing). The downloading of the sources of satellite images algorithm from the sites archives of geodata from the initial data download (the observation area, stage, time period, etc.) is also represented. The procedure of cutting images into segments of varying sizes is described. It allows to use the “contents” of image both in general and for each pixel of the earth's surface to have a whole time series of the spectral brightness coefficients. Examples of time series of geo-referenced source images and the individual indices of the underlying surface are automatically generated by the algorithm method, for specific problems of space debris monitoring (vegetation indices reaction, the degree of soil degradation, detection of indices).