基于遗传算法的相位差异技术图像恢复

Recovering image using a genetic algorithm based phase diversity technology

摘要:

相位差异技术利用焦面图像和固定离焦面图像信息，联合估计波前相位和目标物体.本文论述了相位差异技术的基本理论，由于其评价函数变量多、非线性、无约束条件、计算量非常大的特点提出了将遗传算法运用于相位差异技术.利用计算机开展了大量的模拟仿真实验，仿真实验结果表明，基于遗传算法的相位差异技术可以较准确地恢复光学系统的波前相位信息和目标物体.设计了一个将相位差异技术运用于成像系统的传感器，验证了相位差异技术提高图像清晰度和计算相位波前的能力.

Abstract:

Phase Diversity (PD) technology estimates both the wavefrant phase and object by two focused and defocused images with known phase .The basic principles of wavefront sensing based on phase diversity method are expatiated， and the genetic algorithm is proposed in the data postprocessing of phase diversity testing aiming at dealing with the multi variables， the highly nonlinearity， the Unconstrained condition and large amounts of calculation of merit function.Computer simulations are used to validate the algorithms and techniques， as the results show that this GAbased method can effectively recover wavefront phases and object.An image restoration and wavefront detection sensor by using PD technology was designed.The sensor shows that this Phase Diversity technology can improve image quality .