Automatic Micropipette Tip Detection and Focusing in Industrial Micro-Imaging System

Image processing is basic but important in micromanipulation technology. In this paper, we propose automatic micropipette tip detection and focusing algorithms for an economical and portable industrial micro-imaging system. At present, there are many image processing methods which have obtained good experimental results for microscopic vision. However, there are not suitable image processing methods for images with non-uniform brightness or at different magnification rates. The proposed detection method introduces morphological black hat operator to deal with the non-uniform background and obtains the position of the pipette tip accurately in both clear and blurred images. The proposed focusing method applies a multi-scale gradient transform algorithm to evaluate the clarity of the pipette at different magnifications. A focus strategy is then selected to realize auto-focusing according to the clarity feedback. Experimental results show that the proposed methods obtain better tip detection and clarity evaluation results when the micropipette tip changes from defocusing state to focusing state at different magnifications.

[1]  Clement Leung,et al.  Controlled Aspiration and Positioning of Biological Cells in a Micropipette , 2012, IEEE Transactions on Biomedical Engineering.

[2]  Shaorong Xie,et al.  Robotic Adherent Cell Injection for Characterizing Cell–Cell Communication , 2015, IEEE Transactions on Biomedical Engineering.

[3]  Jun Luo,et al.  Locating End-Effector Tips in Robotic Micromanipulation , 2014, IEEE Transactions on Robotics.

[4]  X Y Liu,et al.  Dynamic evaluation of autofocusing for automated microscopic analysis of blood smear and pap smear , 2007, Journal of microscopy.

[5]  Lu Gui-zhang Batch-Targets-Oriented Auto-Switching of View Field in Micro-Manipulation , 2010 .

[6]  Jesús Chamorro-Martínez,et al.  Diatom autofocusing in brightfield microscopy: a comparative study , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[7]  Edward S. Boyden,et al.  Closed-Loop Real-Time Imaging Enables Fully Automated Cell-Targeted Patch-Clamp Neural Recording In Vivo , 2017, Neuron.

[8]  Yuqian Zhao,et al.  Edge Detection Based on Multi-Structure Elements Morphology , 2006, 2006 6th World Congress on Intelligent Control and Automation.

[9]  Yu Sun,et al.  Automated micro-aspiration of mouse embryo limb bud tissue , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).