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). 4.2 Model Architecture A Convolutional Variational Autoencoder (VAE-CNN) was used to learn latent representations from 9,600 subcrops (80 × 80 × 6) of the study area. The encoder includes three convolutional layers (32, 80, 128 filters) and a dense layer, producing a 3-dimensional latent space via parallel mean and log-variance layers. The decoder mirrors the encoder with transposed convolutions to reconstruct the input (see Table 1). Table 1 – Encoder Architecture Summary Layer Type Output Shape Filters / Units Parameters Input Input Layer 80 × 80 × 6 — 0 Conv1 Conv2D 40 × 40 × 32 32 1,760 Conv2 Conv2D 20 × 20 × 80 80 23,120 Conv3 Conv2D 10 × 10 × 128 128 92,288 Flatten Flatten 12,800 — 0 Dense Fully Connected 128 128 1,638,528 Latent Mean Dense (z_mean) 3 3 387 Latent Log-Var Dense (z_log_var) 3 3 387 Sampling Reparameterization 3 — 0 4.3 Loss Function Figure 16 – Examples of multi-channel input subcrops used in the VAE-CNN model.
RkJQdWJsaXNoZXIy MTM0Mzk2