Fish swarm window selection algorithm based on cell microscopic automatic focus
Fish swarm window selection algorithm based on cell microscopic automatic focus
复制标题
DOI：
10.1007/s10586-017-0752-4
复制
发表时间：
2017-03
期刊：
影响因子：
--
通讯作者：
Fengshou Zhang;Siwen Li;Zhi-gang Hu;Zhe Du
中科院分区：
文献类型：
--
作者：
Fengshou Zhang;Siwen Li;Zhi-gang Hu;Zhe Du
传统的自动聚焦窗口选择算法的选择窗口主要集中在图像的中心，因此随机分布的细胞总是失焦。针对这一问题，在分析不同聚焦窗口选择算法性能的基础上，提出了一种在传统鱼群算法的基础上改进的自动聚焦窗口算法：鱼群窗口选择算法。通过对传统鱼群算法和改进鱼群算法获得的焦点窗口内图像进行对比分析，可以得出改进算法的焦点窗口可以包含更多的细胞和目标体。具体而言，由于采用鱼群窗口选择算法，选择窗口内高频图像的数量大大增加，最优解收敛于0.999，并且改进算法的精度高、聚焦精度高，所获得的显微细胞图像的清晰度估计值也有所提高。
The selection window of selection algorithm used in traditional automatic focus window concentrates mainly on the center of image, so the randomly distributed cells would always be out of focus. To address this problem, on the basis of analyzing the performance of selection algorithm of different focusing windows, a modified auto-focus window algorithm upon traditional fish swarm algorithm has been proposed: fish swarm window selection algorithm. After comparatively analyzing the images in focus window that are obtained by traditional and improved fish swarm algorithm, a conclude can be drawn that the focus window of modified algorithm can contain more cells and target bodies. To be specific, owing to fish-swarm window selection algorithm, in the selection window the quantity of the high frequency of images greatly increases, the optimal solution converges to 0.999, and the estimated value of sharpness of the obtained microscopic cell images also improves with high precision and high accuracy of focus of the improved algorithm.