ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., III-4, 3-10, 2016
https://doi.org/10.5194/isprs-annals-III-4-3-2016
© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.
 
03 Jun 2016
TRUE ORTHO GENERATION OF URBAN AREA USING HIGH RESOLUTION AERIAL PHOTOS
Yong Hu, David Stanley, and Yubin Xin PCI Geomatics, 90 Allstate Parkway, Markham, Canada
Keywords: Orthorectification, Elevation buffer, Visibility Analysis, Segment, Occlusion Compensation Abstract. The pros and cons of existing methods for true ortho generation are analyzed based on a critical literature review for its two major processing stages: visibility analysis and occlusion compensation. They process frame and pushbroom images using different algorithms for visibility analysis due to the need of perspective centers used by the z-buffer (or alike) techniques. For occlusion compensation, the pixel-based approach likely results in excessive seamlines in the ortho-rectified images due to the use of a quality measure on the pixel-by-pixel rating basis. In this paper, we proposed innovative solutions to tackle the aforementioned problems. For visibility analysis, an elevation buffer technique is introduced to employ the plain elevations instead of the distances from perspective centers by z-buffer, and has the advantage of sensor independency. A segment oriented strategy is developed to evaluate a plain cost measure per segment for occlusion compensation instead of the tedious quality rating per pixel. The cost measure directly evaluates the imaging geometry characteristics in ground space, and is also sensor independent. Experimental results are demonstrated using aerial photos acquired by UltraCam camera.
Conference paper (PDF, 969 KB)


Citation: Hu, Y., Stanley, D., and Xin, Y.: TRUE ORTHO GENERATION OF URBAN AREA USING HIGH RESOLUTION AERIAL PHOTOS, ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., III-4, 3-10, https://doi.org/10.5194/isprs-annals-III-4-3-2016, 2016.

BibTeX EndNote Reference Manager XML