Image Processing Technique for Characteristic Test of Cell Based on Electrorotation Chip

On the basis of image characteristic analysis of electrorotation (ROT) chip video, the initial cell contours of various movement objects in each frame have been segmented and each cell contour has been purified to extract using the Snake model, and then the projected area and centroid integral of cell etc. have been calculated. Moreover, the Snake cell contour is fitted into an ellipse using the least squares ellipse fitting algorithm. Through omitting the objects whose area, centroid and deflection angle change suddenly with the statistical rule, then the average of the direction (i.e. ellipse deflection angle) difference of the surplus objects between the continual two frames is taken as the rotation vector of the whole cells; The multi-objects are synchronously tracked in each continual dual- frames by the frame sequence. Finally the actual rotation speed of each cell is calculated at a specified equal time interval.