| الوصف: |
An essential component of remote sensing, image analysis, and patternrecognition is image categorization. The classification of land use usingremotely sensed data creates a map-like representation as the final form ofthe investigation. With its ability to effectively categorize satellite images,machine learning (ML) algorithms have gained significant traction in anumber of fields, including land-use planning, disaster response, and naturalresource management. Ensemble learning is also a widely used technique forenhancing the precision of satellite image categorization, which combinesmultiple models to get more precise predictions. Holdout is an ensembletechnique, where multiple ML algorithms are used for training on the samedataset. The primary goal of this study is to create a holdout model forclassifying satellite images. Initially, this study explores the usage of MLalgorithms namely support vector machines (SVM), k-nearest neighbor(KNN), decision trees (DT), gradient boosting classifier (GBC), histogram-based GBC (HGBC), random forest classifier (RF), bagging classifier (BC),XGBoost classifier for classifying satellite images. Later, GBC, HGBC, RF,BC, and XGBoost are combined to build a stacking model. The baggingensemble model outperforms all other methods and reaches an accuracy of88.90%. Finally, blending models with holdout approach were developedand achieved accuracy of 93.70%, 94.14%, and 93.87% which outperformedall previous algorithms. |