Exploration of boron mine sites with remote sensing and machine learning interaction

Objects respond differently to electromagnetic radiation according to their physical and chemical compositions. Radiometric sensors of satellites record electromagnetic reflection values obtained from the earth's surface at various wavelengths. Remote sensing is used to distinguish, identify and analyze objects with different reflection values. The study aims to develop boron mineral exploration methods with the interaction of remote sensing and machine learning. Boron, a mineral of strategic importance, is the focus of this study. Using Google Earth Engine, satellite images obtained at different dates were processed and analyzed in the cloud environment. With this approach, large data sets can be processed and analyzed quickly, and an application that classifies boron minerals according to parameters such as time, satellite name and machine learning method has been developed. The study shows that remote sensing and machine learning methods can be used effectively in determining boron formations on earth. Patterns indicating the presence of boron were determined with machine learning algorithms, ensuring accurate results. Overall, the interaction of remote sensing with machine learning offers significant improvements in the efficiency and accuracy of mineral exploration. This research underscores the potential of modern technological tools in enhancing mineral exploration processes, ultimately contributing to better resource management and utilization strategies. Keywords: Remote Sensing, Google Earth Engine, Machine Learning, Boron