Abstract: In view of the new task of extracting canopy cover in national geographic condition monitoring work, through in-depth study and analysis of spectral and texture characteristics of main ground features in housing construction area, the normalized vegetation index (NDVI), contrast and entropy are used as judgment rules. According to the idea of object-oriented, an extraction method of canopy coverage in building area based on high-resolution remote sensing image is designed. The experimental results show that the method can automatically extract the canopy coverage of the building area, greatly reduce the workload of the current common visual interpretation methods, and can effectively distinguish other ground features in the building area with similar spectral characteristics of canopy coverage compared with using a single image feature.

Key words: high-resolution remote sensing, geographical conditions monitoring, building area, canopy coverage, texture features, spectral characteristics, object-oriented classification