Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-A new study on support vector machines is now available. According to news originating from Kerman, Iran, by NewsRx co rrespondents, research stated, "This study aims to map the non-sulphide Zinc (Zn )-bearing gossans at the Gujer Zn deposit area, Central Iran, using Landsat 8 Op erational Land Imager (OLI) and GeoEye-1 satellites. The colour composites, Prin cipal Component Analysis (PCA), and Support Vector Machine (SVM) were adopted fo r image analysis." The news journalists obtained a quote from the research from Graduate University of Advanced Technology: "Zn-bearing gossans contain Fe-oxyhydroxide minerals di splaying spectral characteristics in visible and infrared (IR) wavelengths. The application of colour composites using GeoEye-1 images resulted in the delineati on of gossans (real target) and ferruginous sandstones (false targets) having th e same colour tone in the study area. IR spectroscopy of ore samples showed that hemimorphite exhibits low absorption in shortwave infrared (SWIR) wavelengths. Consequently, the Crosta-PC analysis was conducted using bands 4, 5, SWIR-1, and SWIR-2 of Landsat OLI to enhance only ore gossans. Five target zones were speci fied using the Crosta technique. The SVM method was performed to increase the ac curacy of image analysis using the Radial Basis Function (RBF) kernel. The SVM-R BF method accomplished enhancing ore gossans by defining a new target zone. Acco rding to the results, the application of the Crosta technique using bands 4, 5, SWIR-1, and SWIR-2 of Landsat OLI can specify ore gossans and eliminate the inte rfering effect of ferruginous sandstones in similar geological settings."
NETL
NSTL