Nature and artificial grassland classification for urban area

Using remote sensing to monitoring grassland for urban area is one of the major environmental applications today. Experts urge that there should be more nature grassland than artificial grassland in urban area, because the former is much important than the latter. It is difficult to identify them using normal methods such as pixel-based classification because both grasslands have similar spectral information. Going far beyond the methodical limits of pixel-based and manual interpretation approaches; object-oriented image analysis approaches are used for extracting information. This paper presents a snapshot of work to detect different grassland information in Daqing city using this patented technique. The result of grassland information extraction is promising and the precision of classification is higher than other conventional processes. It is obvious that this new image analysis approach offers a satisfying solution to extract information quickly and efficiently.