Machine learning-based vegetation cover analysis in Lake Hawdon North using point segmentation

This study presents a comprehensive methodology and validation of a machine learning (ML) based approach for vegetation cover analysis using high-resolution RGB imagery in Lake Hawdon North. By employing convolutional neural networks like DeepLabV3 [1], the project demonstrates a precise classification of various vegetation types, followed by a rigorous validation process. The results showcase the efficacy of point annotations and ML in remote sensing and geospatial analytics.