RESUMO
In this paper, GLUT4 vesicles are observed in real-time under TIRF microscopy and a new three-dimensional single particle tracking algorithm according to the unique features of TIRF is put forward. Firstly a fluorescence correction procedure was processed to solve the problem of fluorescence bleaching over time and mobile vesicles were segmented by an adaptive background subtraction method. Kalman filtering was then introduced to track the granules so as to reduce the searching range and to avoid the disturbance of background noise and false targets. In the experiments the algorithm was applied in analyzing the long-distance movement of GLUT4 vesicles. The experimental results indicate that the algorithm has achieved robust tracking of the vesicles in the imaging plane and has effectively calculated the position in the direction orthogonal to the imaging plane.