混入逃逸函数的遗传算法优化超大视场光学系统

超大视场光学成像系统在各领域的应用越来越多,但却缺少能够对该类光学系统的像差进行参量化设计的方法。将遗传算法和逃逸函数相结合对超大视场光学系统进行了优化设计。首先,修正了基于平面对称像差理论的超大视场光学系统的评价函数;然后针对遗传算法在优化超多参量光学系统时,其优化解的鲁棒性较差的问题,采用在遗传算法中混入逃逸函数来改善算法的鲁棒性。最后应用改进的算法分别对鱼眼镜头和折反射全景成像系统进行了优化计算,结果表明,优化后光学系统的像质比参考设计有较大的改善。

Application of the genetic algorithm mixed with escape function to optimization of the ultra-wide-angle optical systems

Ultra wide angle imaging system is applied in more and more fields. However, there is no method of parametric design of the aberrations of the ultra-wide-angle. Therefore, a genetic algorithm mixed with escape function is utilized to optimize ultra-wide-angle optical systems. First, the evaluation function of the ultra-wide-angle optical systems based on wavefront aberration theory is modified. A genetic algorithm mixed with escape function(MEFGA)is proposed to solve the poor robustness of the genetic algorithm when it is applied to optimize the optical systems of ultra many parameters. Finally, this improved genetic algorithm is used to optimize a fisheye lens and catadioptric panoramic optical system. The calculation results show that the image quality of the resultant optical systems is significantly improved.