Riparian zones consist of important environmental regions, specifically to maintain the quality of water resources. Accurately mapping forest vegetation in riparian zones is an important issue, since it may provide information about numerous surface processes that occur in these areas. Recently, machine learning algorithms have gained attention as an innovative approach to extract information from remote sensing imagery, including to support the mapping task of vegetation areas. Nonetheless, studies related to machine learning application for forest vegetation mapping in the riparian zones exclusively is still limited. Therefore, this paper presents a framework for forest vegetation mapping in riparian zones based on machine learning models using orbital multispectral images. A total of 14 Sentinel-2 images registered throughout the year, covering a large riparian zone of a portion of a wide river in the Pontal do Paranapanema region, São Paulo state, Brazil, was adopted as the dataset. This area is mainly composed of the Atlantic Biome vegetation, and it is near to the last primary fragment of its biome, being an important region from the environmental planning point of view. We compared the performance of multiple machine learning algorithms like decision tree (DT), random forest (RF), support vector machine (SVM), and normal Bayes (NB). We evaluated different dates and locations with all models. Our results demonstrated that the DT learner has, overall, the highest accuracy in this task. The DT algorithm also showed high accuracy when applied on different dates and in the riparian zone of another river. We conclude that the proposed approach is appropriated to accurately map forest vegetation in riparian zones, including temporal context.

Mapping urban tree species supports green infrastructure planning. An essential issue refers to the monitoring of exotic species that may become invasive. Street-level imagery provides a complementary perspective to aerial images for species identification that are difficult to distinguish from above. In this context, our study aimed to evaluate deep learning-based object detection and image segmentation approaches to identify a potentially invasive tree species known as Leucaena leucocephala in an urban environment in Brazil, using 422 street-level images acquired from Google Street View (SV) and mobile phones (MPs). Object detection models (YOLOv8 and DETR) and a foundation segmentation model (SAM, zero-shot) were applied to assess how deep learning paradigms perform under heterogeneous urban imaging conditions. YOLOv8 achieved detection performance with mAP50 above 0.83 and recall up to 0.76. DETR showed domain sensitivity, with mAP50 of 0.45 in SV images and 0.84 in MP imagery. For segmentation, SAM zero-shot achieved 0.92 accuracy and 0.93 F1-score in SV images, decreasing to 0.63 accuracy and 0.66 F1-score in MP images. Overall, this study demonstrates that combining detection and segmentation approaches provides complementary information for urban vegetation monitoring, supporting decision-making related to invasive species management and sustainable urban landscape planning.