细胞显微操作的自动调焦及图像处理

Auto-focusing based location method and image processing algorithm for a robot-assist embryo microinjection system

随着现代生物医学技术研究的不断深入,研制高精度、高效率的细胞操作机器人已成为国际学术界和工业界的焦点。本文以优化显微操作机器人的操纵空间和操作流程为目的,应用自调焦图像算法评价、机械手运动控制和主从机通信等技术实现了基于自动调焦定位的细胞操作方案。首先,以斑马鱼卵细胞为例,结合一类球形卵细胞的对称性及细胞定位操纵中定位特点,对不同聚焦函数进行评价及图像处理算法优化,提出适宜细胞操纵的自动调焦定位方法。其次,以精确度和快速性为依据,对不同图像分割与识别算法进行评价,提出最适宜细胞和注射针的自动化识别及定位算法。最后,通过RS-232串口通讯,实现了上位机对机械手自动化操纵控制。试验结果表明,提出的自调焦定位操纵策略运行稳定、精度高,满足细胞显微操作机器人的高精度及自动化操作的要求。

With the development of the biomedical technology, researches on the high precision and high efficient cell manipulation robots have attracted more and more interests from academic and industry researchers. In order to reduce the sophistication of the whole manipulation and optimize the manipulation procedure, a vision servoing method was adopted by integrating the auto- focusing image process, motion control and communication technologies between the host computer and robot. Firstly, consider- ing the manipulation scenario, a number of autofocus functions were reviewed and optimized. Then, different imaging segmentation and detection methods were evaluated from the aspects of precision and fastness. Finally, based on the imaging processing results, RS-232 series communication was employed to realize the automatic control of cell manipulator. Experimental results showed that the proposed auto-focusing detection strategy is robust and with high precision, which could meet the requirement of automatic cell manipulation.