Preface: Smart solutions in mathematical engineering and sciences theory

Abstract

The role of this special issue is mainly focused on remote sensing and machine learning applications for environments. The rapidly increasing availability of multispectral, high spatial resolution imagery, collected by satellites, cubesats, and airborne sensors, presents an opportunity to detect landscape change, landuse landcover, surface temperature, nature disaster with increased spatial detail of research environment and applications. The research environment studies utilizing data from several satellite imageries such as LANDSAT, WORLDVIEW, SPOT, LIDAR data, Sentinel and MODIS other satellite imagery. Today, a new generation of research environment studies using several satellite imagery with different spatial resolution based on research study through the capitalizing on the availability of data from high spatial resolution global monitoring missions. For example, the unprecedented 45-year long global Landsat archive is increasingly used to analyze past and present global land and water changes, and higher temporal frequency global observations from Sentinel are enabling the use of dense high-resolution time series for near real time monitoring. In addition to Sentinel and Landsat, data from other global Landsat-class missions are increasingly being integrated into virtual Earth observation constellations that further advance global land and water monitoring. These challenges all point to the need for improved image processing approaches specific to multispectral, high spatial resolution imagery. In this Special Issue, the methodological contributions in terms of novel machine learning algorithms as well as the application of innovative techniques to relevant scenarios from hyperspectral data. On the other hand, the environmental modelling can be described as a simplified form of a real system that enhances our knowledge of how a system operates. Such models represent the functioning of various processes of the environment, such as: processes related to atmosphere, hydrology and land surface, among others. In fact, environmental models may span a wide spectrum of geographic (i.e., from local to regional to global-levels) and temporal (i.e., diurnal to monthly to annual to decadal-levels) scale. They often integrate various aspects of the environment that can be described upon employing various types of models, such as process-driven, empirical or data-driven, deterministic, stochastic, etc