Abstract

Machine learning-based Support vector machine (SVM) and Extended Local Binary Patterns (ELBP) algorithms were used to classify satellite images into 24 different categories. In addition to satellite photos, this study can classify 24 other classifications. However, identifying the traits of those other classes, such as the human face, football, and rugby, is similarly simple because these other classes contain some distinguishing characteristics that allow for easy classification. The underlying difficulty with satellite photographs is that multiple satellite images may have different properties, making classification difficult. Another issue is that most satellite images are distorted by noise.

Description