A novel 3D stitching method for WLI based large range surface topography measurement
| Autor: | Xiaojun Liu, Lei Zili, Wenlong Lu, Liangzhou Chen, Tengfei Yuan, Qian Li, Zhao Li |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
business.industry
Noise (signal processing) Computer science ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Process (computing) 02 engineering and technology 01 natural sciences Signal Atomic and Molecular Physics and Optics Electronic Optical and Magnetic Materials 010309 optics Image stitching Interferometry Matrix (mathematics) Optics Interference (communication) 0103 physical sciences 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Electrical and Electronic Engineering Physical and Theoretical Chemistry business |
| Zdroj: | Optics Communications. 359:435-447 |
| ISSN: | 0030-4018 |
| Popis: | 3D image stitching is an important technique for large range surface topography measurement in White-Light Interferometry (WLI). However, the stitching accuracy is inevitably influenced by noise. To solve this problem, a novel method for 3D image stitching is proposed in this paper. In this method, based on noise mechanism analysis in WLI measurement, a new definition of noise in 3D image is given by an evaluation model for difference between the practical WLI interference signal and the ideal signal. By this new definition, actual noises in 3D image are identified while those practical singular heights on surface will not be wrongly attributed to noise. With the definition, a binary matrix for noise mark corresponding to 3D image is obtained. Then, the matrix is devoted, as an important component, to establish a series of new algorithms of capability for suppressing the adverse effects of noises in each process of the proposed stitching method. By this method, the influence of the noises on stitching is substantially reduced and the stitching accuracy is improved. Through 3D image stitching experiments with noises in WLI, effectiveness of the proposed method is verified. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect