Remote Sensing Application in Forest Monitoring: An Object Based Approach

Abstract

The objective of this study was to establish forest map in 2015 using object-based classification technique in SPOT 6 image and analyze land use/land cover changes in landscape of Yen Nhan commune, ThanhHoa province in Vietnam over a period of 15 yeas (2000-2015). Object-based methods allow integration of different object features, such as spectral values, shape, and texture. One of its strength is the ability to combine spectral information and spatial information for extracting target objects. Few studies have explored the application of object-based approaches to classify forest. This paper introduced an object based method to SPOT6 image to map the land cover in Yen Nhan commune in 2015. This approach applied multi-resolution segmentation algorithm of eCognition Developer and an object based classification framework. In addition, existed forest maps from 2000 to 2015 were used to analyze the change in forest cover in each 5 years period. The object based method clearly discriminated the different land cover classes in Yen Nhan. The overall kappa value 0.73 was achieved. The estimation of forest area was 89.05 % of total area in 2015. By overlaying achieve forest maps of 2000, 2005, 2010, the classified map of 2015 showed vegetation changed remarkably during 2000-2015.