本文总结了作者过去两年中对“数字成像系统的自动对焦算法”进行研究的成果。
This paper summarizes the author's research accomplishments on the project of "Auto-focusing Algorithm in Digital Imaging Systems" in the last two years.
针对显微镜下两层物体的大规模图像采集问题，提出了基于图像边缘和邻域预测快速自动对焦算法。
Aiming at the problem of large-scale microscopy images acquisition of two-layer object, an auto-focusing algorithm was proposed based on image edge and adjacent area forecasting.
自动对焦的过程就是运用爬山搜索算法求取对焦评价函数最大值的过程。
The procedure of auto-focusing is to getting the maximum of the focusing (evaluation) function employing the mountain-climb searching algorithm.
重新设计的算法，使得它有可能加快，约减少自动对焦时间。在广角端，其中23%的速度最快的一流水平。
The redesigned algorithm makes it possible to speed up the AF time by reducing it approx. 23% at wide-end, which is the fastest level of speed in class.
对数字成像系统自动对焦的反馈控制算法和软硬件结构进行了讨论。
The feedback control hi auto-focus and the hardware and software architecture are also discussed.
在软件上，主要是搜索算法的优化，还有对焦评价函数和对焦窗口的选择，以及相对于硬件上优化后所采取的一些改进。
About software, searching algorithm has been optimized. The focus evaluation function and the choice of the focusing window have also been described.
现有的许多显著目标检测算法，大都依赖像素间的相互关系，而缺乏对焦点对象特征的理解。
Many existing detecting algorithms of salient object mostly depend on the relationship between the pixels and lack the comprehension of salient object features.
针对焦炉温度的控制特点和难点，提出了模糊免疫自适应pid控制算法并对它进行了详细的分析设计。
According to the characteristics and difficulties of the coke oven control, a fuzzy immune adaptive PID control method is brought forward and designed in detail.
针对焦炉集气管压力系统具有多回路、强耦合、非线性的特点，提出了一种综合控制算法。
According to the characteristics of nonlinear, close coupling and multiple loop for coke oven collector pressure control systems, a new mixed control method is proposed.
传统的逐点重建法算法简单，但实验装置复杂，且存在对焦误差。
The traditional point-by-point imaging modality is very simple, however, it needs complex experimental setup and there exist positioning error.
实验结果表明：将所提出的对焦评价算子和搜索算法结合使用， 能够准确实现自动聚焦，准确度达到95%以上。
The results show that the correlation operator with the cooperation of the lean operator can locate the prime focus position accurately, the accuracy is over 95%.
实验结果表明：将所提出的对焦评价算子和搜索算法结合使用， 能够准确实现自动聚焦，准确度达到95%以上。
The results show that the correlation operator with the cooperation of the lean operator can locate the prime focus position accurately, the accuracy is over 95%.
应用推荐