Mangrove forests are useful in protecting coastal lines, reducing atmospheric carbon dioxide (CO2), and providing shelters and foods for ecosystems, but they are among the top threatened habitats worldwide caused by human activities, climate change, and coastal-related natural disasters such as tsunami, flood, storm, etc. Understanding quantitative and spatial information of mangrove associations is essential for managing mangrove forests effectively. The main aim of this research is to investigate the changes of mangrove forest associations in tropical regions and possible reasons of changes via the case study of the Can Gio biosphere reserve in Viet Nam by using remote sensing datasets (SPOT 4 and 5 imagery), object-based image analysis and Support Vector Machine classifier. The classification results showed that the highest overall accuracy of this combination was 81.6%, given 420 reference objects, confirming the applicability of using remote sensing on assessing the development of mangrove forests. For the case study, we found that there was a decrease of the Avicennia alba – Sonneratia alba (Association I) area by 20.1% caused by the development of aquaculture and other human activities, but there was an increase of the Rhizophora apiculata (Association II) area by 34.8%. These trends may result in the exacerbated risk of soil erosion and make the region become more vulnerable to climate change and tropical storm events. Management policies should be revised to harmonise the benefits of economic development, environmental protection, and species diversity.