145 3.3. Preprocessing of Layers First, all raster layers were spatially aligned and resampled to a common grid with a spatial resolution of 100 m. This step ensured pixel-wise correspondence across all input features. Missing and NoData values were addressed using a K-Nearest Neighbors (KNN) imputation approach with k = 10, where missing values were replaced using the median of neighboring pixels. Finally, all features were normalized using MinMax scaling, transforming values into a [0,1] range. 4. METHODOLOGY 4.1. Data Preparation A sliding window strategy was applied to systematically extract spatial samples from the study area. Windows of 100 × 100 pixels were generated using a stride of 50 pixels, ensuring overlapping coverage and improved spatial representation. From each window, four random 80 × 80 subcrops were generated through a jittering process. This approach increases data variability while preserving local geological patterns, allowing the model to learn robust spatial features. This procedure resulted in approximately 9,600 samples, providing comprehensive coverage of the study area (See Figure 4).
RkJQdWJsaXNoZXIy MTM0Mzk2