<p>Mapping coastal landforms is essential for understanding shoreline dynamics and environmental change and supporting sustainable coastal zone management. This study integrates remote sensing and machine learning techniques to improve the classification of coastal geomorphological features. Multiple image processing methods – Modified Enhanced Texture Filtering with Clustering (METF-C), median filtering, Gaussian filtering, and image pyramids – were evaluated to determine their effectiveness in preserving key coastal features. Among these, METF-C demonstrated superior performance in retaining geomorphological detail, maintaining edge sharpness, and enhancing texture clarity. Gaussian filtering exhibited high information retention but failed to preserve texture effectively. Image pyramids maintained structural integrity but showed lower feature retention efficiency. Median filtering offered moderate edge preservation but was less effective overall. The analysis indicates that METF-C is the most effective technique for detailed coastal mapping, particularly in complex and dynamic environments. By combining these advanced image processing techniques with machine learning classification algorithms, the study achieves enhanced detection and interpretation of coastal landforms. This approach supports more accurate monitoring of shoreline changes and erosion patterns, ultimately contributing to more informed coastal planning and management. The findings highlight the potential of integrating machine learning with refined image processing workflows for automated, large-scale coastal geomorphology mapping and environmental monitoring.</p>