Rapid and accurate phase matching for 3D measurement

Abstract

A rapid and accurate matching method based on absolute phase maps is proposed in this paper. This algorithm is composed of the matching mask, the optimal interpolation, the filter of the parallax and the smoothing of the point cloud. First, the average intensity image and the co-occurrence matrix based on the gray and modulation levels are used to obtain the object mask. Then, a novel weighted interpolation is adopted to get the sub-pixel parallax. Next, a filter is used to remove the solitary points and smooth the parallax between intervals. After this operation, the parallax becomes smoother. Finally, a Gaussian smoothing is applied on the point cloud to reduce the burr within an interval. Experiments show that the proposed method can acquire the rebuilding region and the point cloud rapidly and accurately.