光学与光电技术, 2019, 17 (4): 77, 网络出版: 2019-09-27
干涉检测中常用解包裹算法噪声处理能力比较
Comparison on the Noise Processing Ability of Common Unwrapping Algorithms in Interference Detection
干涉检测 相位解包裹 离散余弦变换 噪声 欠采样 interference detection phase unwrapping discrete cosine transform noise under sampling
摘要
选择合理的相位解包裹算法是提高检测精度的重要保证, 由于检测过程噪声的影响不可避免, 因此需要在数据处理过程中考虑到算法的噪声处理能力。为了能准确地处理含强噪声的包裹相位图, 从最小二乘解包裹算法的原理出发, 具体分析了基于离散余弦变换的最小二乘法(DCT-LS)及在其基础上发展起来的三种优化算法的特点, 并对不同噪声条件下的解包裹计算结果进行了对比研究。仿真结果表明: 四种算法针对不同类型的噪声处理效果各异, 比较而言LS-LS抗噪能力最强, 在强噪声情况下解包裹结果的残差值仅为1.171 8 rad, 远小于其他三种算法, 且唯一具有较好的抗欠采样能力。
Abstract
Choosing a reasonable phase unwrapping algorithm is the guarantee to high accuracy of optical test. Since the influence of noise in the detection process is unavoidable, it is necessary to consider the noise processing ability of the algorithm in the process of data processing. In order to process the wrapped phase map with strong noise, based on the principle of least squares unwrapping algorithm, the characteristics of the least squares method of discrete cosine transform (DCT-LS) and other three optimization algorithms are analyzed. Furthermore, the results of unwrapping calculation under different noise conditions are compared. The results show that the four algorithms have different effects on different types of noise processing. In comparison, LS-LS has the best anti-noise ability. The residual value of the unwrapping result in the case of strong noise is only 1.171 8 rad, which is much smaller than the other three algorithms, and the only one has good anti-under-sampling capability.
彭景, 张蓉竹. 干涉检测中常用解包裹算法噪声处理能力比较[J]. 光学与光电技术, 2019, 17(4): 77. PENG Jing, ZHANG Rong-zhu. Comparison on the Noise Processing Ability of Common Unwrapping Algorithms in Interference Detection[J]. OPTICS & OPTOELECTRONIC TECHNOLOGY, 2019, 17(4): 77.