Passive depth estimation using stereo vision, an experimental study
MetadataShow full item record
This master s thesis is an experimental study on passive stereo techniques forretrieving 3D scene information. The stereo camera system is tested for findingthe position of a platform relative to the stereo camera where the measurementsare used for navigating an autonomous docking system installed on offshore vessels. The accuracy of passive stereo techniques depends on having precise knowledge of the cameras position, orientation and internal parameters. Calibratingthese parameters is a key part of the success of the stereo camera. For a pointin the scene imaged from two different known view points, the displacement ofthe point reprojection in the images are inversely proportional to depth and canb e used to compute the 3D coordinate. The problem of establishing correspondences in the image pair is a difficult task and there exist several algorithmsfor solving the correspondence problem.In this thesis, we have evaluated the accuracy of the stereo camera systemfor computing 3D information of the scene and proposed an optimal baselinelength to improve the accuracy. Six different of algorithms have been tested forsolving the correspondence problem for this application. The algorithms considered are the correlation based methods: Sum of absolute differences, squaredsum of differences and normalized cross correlation. Together with the featurebased methods: Harris corner detector, scale invariant feature transform andspeeded up robust features. The performance of each algorithm is comparedthrough a field experiment to conclude on the b est one.Based on the results of the experiment, the correlation based methods provesto be best suited for this application where the sum of squared differences isconsidered as the superior algorithm for solving the correspondence problem.