UAV remote sensing (UAV-RS) has been applied in many fields, due to its advantages of high time and higt-resolution. but recent research is mainly focused on two-dimensional application such as image regiistration, mosaic, classification, etc. creatively, this study focuses on using UAV images for three-dimensional application. firstly, SIFT (scale-invariant feature transform) algorithm is employed to detect correspondence points between two adjacent images automatically, and then multi-view geometry and the theory of binocular stereo vision are used for matching for image and for 3-D stereo processing of image, finally the two images are synthesized for enhanced display with three-dimensional display devices. developed in VC program environment, this study builds its prototype system, which achieves a nice three-dimensional visualization under any of the three modes including anaglyphic 3D, polarization 3D and active shutter 3D. the result shows that 3-D stereo processingbrings structured details. which provides a new way for feature interpretation. |