Abstract
This paper presents an efficient and robust phase unwrapping algorithm which combines an unscented Kalman filter (UKF) with a strategy of quantizing a paths-guided map and a pixel classification strategy based on phase quality information. The advantages of the proposed method depend on the following contributions: (1) the strategy of quantizing the paths-guided map can accelerate the process of searching unwrapping paths and greatly reducing time consumption on the unwrapping procedure; (2) the pixel classification strategy proposed by this paper can reduce the error propagation effect by decreasing the amounts of pixels with equal quantized paths-guided value in the process of unwrapping; and (3) the unscented Kalman filter enables simultaneous filtering and unwrapping without the information loss caused by linearization of a nonlinear model. In addition, a new paths-guided map derived from a phase quality map is inserted into the strategy of quantizing the paths-guided map to provide a more robust path of unwrapping, and then ensures better unwrapping results. Results obtained from synthetic data and real data show that the proposed method can efficiently obtain better solutions with respect to some of the most used algorithms.
© 2015 Optical Society of America
Full Article | PDF ArticleMore Like This
Xianming Xie
Opt. Express 24(17) 18872-18897 (2016)
XianMing Xie and YingHui Li
Appl. Opt. 53(18) 4049-4060 (2014)
Xianming Xie and Gaoxing Dai
Appl. Opt. 56(34) 9423-9434 (2017)