The extraction of wetland vegetation information based on UAV remote sensing images

Unmanned aerial vehicle (UAV) have been increasingly used for natural resource applications in recent years as a result of their greater availability, the miniaturization of sensors, and the ability to deploy UAV relatively quickly and repeatedly at low altitudes. In this paper, the wetland vegetation information is extracted from UAV remote sensing images by object-oriented approach. Firstly, the images are segmented and images object are build. Secondly, VDVI, VDWI, spectral information and object geometry information of images objects are comprehensively applied to build wetland vegetation extraction knowledge base. Thirdly, the results of wetland vegetation extraction are improved and completed. The results show that better accuracy of wetland vegetation extraction can be obtained by the proposed method, in contrast to the pixel-oriented method. In this study, the overall accuracy of classified image is 0.968 and Kappa accuracy is 0.934.