A Data Analytic Model for Remote Sensing Data Using Data Mining Techniques

Abstract

"Big Data" is a term used to describe the massive quantity of real-time data generated by the digital world's remote sensing assets. Data from remote sensing is far more complex than it appears at first glance; thus, retrieving relevant information in an effective manner can lead a system to severe computing issues, such as analysis and storage wherever data are remotely acquired. Accordingly, a design specification that allows for both real-time and offline data processing is needed considering the foregoing considerations. So, in this research, we offer a real-time data mining framework for remote sensing satellites. Ubiquitous Big Data collecting, processing, and decision-making units are the three key components of the suggested architecture. Using the suggested architecture, only valuable data will be divided up, balanced out, and processed in parallel. Consequently, real-time remotely sensed Big Data employing Earth observatory systems may be efficiently analysed using data mining approaches. As an additional benefit, the suggested architecture includes the capacity of retaining raw data to undertake offline analysis on huge dumps, if necessary. Big data from Earth observation satellites are analysed using Hadoop for land and water areas.