散斑干涉相关条纹的交叉熵方法研究

现代电子散斑干涉计量过程中，状态变化信息的相关条纹表征对后续待测信息的提取起着重要作用，直接影响到后续处理的难度和信息提取的质量。本文根据交叉熵的基本概念和散斑干涉计量过程状态变化信息表征基本原理的分析，提出了一种新的相关条纹图生成方法－交叉熵法，通过详细的理论推导和应用效果的定量分析表明交叉熵方法所产生的相关条纹与减相关条纹在本质上是一致的，提出的方法完全有效。同时，由于交叉熵条纹表征方法融入了自适应同态滤波，与减条纹相比具有更好的条纹图对比度、较低的散斑指数和更好的包裹相位信息提取质量。

Correlation Fringe of Speckle Pattern Interferometry Based on Cross-entropy

In the modern electronic speckle pattern interferometry, the correlation fringe representation for the state change plays an important role in extracting the messages and the difficulty and precision of extracting and has a direct impact on. A novel approach named cross-entropy correlation fringe was presented in this paper, which was based on the concept of cross-entropy and the basic principles of the fringe representation for the state change in speckle pattern interferometry. After the theory analysis and applications, it was indicated that the correlation fringe generated by cross-entropy are essentially coincident with the one generated by subtraction. The homomorphic filtering was integrated into the generation of cross-entropy fringe inherently, which promises a better performances relate to the subtraction fringe. The cross-entropy fringes have a better performance in the fringe contrast and lower speckle index, and have a better quality in the wrapped phase map. The cross-entropy fringe presented in this paper is effective.