Title :
Applying computer vision techniques to perform semi-automated analytical photogrammetry
Author :
Nilosek, David ; Salvaggio, Carl
Author_Institution :
Chester F. Carlson Center for Imaging Sci., Rochester Inst. of Technol., Rochester, MA, USA
Abstract :
The purpose of this research is to show how common computer vision techniques can be implemented in such a way that it is possible to automate the process of analytical photogram-metry. This work develops a workflow that generates a sparse three-dimensional point cloud from a bundle of images using SIFT, RANSAC, and a sparse bundle adjustment along with basic photogrammetric methods. It then goes on to show how the output of the sparse reconstruction method can be used to generate denser three-dimensional point clouds that can be facetized and turned into high resolution three-dimensional models. This workflow was successfully tested on a five image dataset taken with RIT´s WASP imaging sensor over the Van Lare wastewater treatment plant in Rochester, NY.
Keywords :
computer vision; geophysical image processing; image reconstruction; image sensors; photogrammetry; transforms; wastewater treatment; RANSAC; RIT WASP imaging sensor; SIFT; Van Lare wastewater treatment plant; computer vision techniques; random sample consensus; scale invariant feature transform; semiautomated analytical photogrammetry; sparse bundle adjustment; sparse reconstruction method; sparse three-dimensional point cloud; Clouds; Computational modeling; Computer vision; Feature extraction; Mathematical model; Solid modeling; Sparse matrices;
Conference_Titel :
Image Processing Workshop (WNYIPW), 2010 Western New York
Conference_Location :
Rochester, NY
Print_ISBN :
978-1-4244-9298-5
DOI :
10.1109/WNYIPW.2010.5649777
