This paper proposed the “self-checking” algorithm to improve the detection accuracy of multiple moving targets in time-series fluorescence images, such as vesicles. The main idea of this algorithm is to construct a multi-kernel function superposition model and use the model to fit the data at the indistinguishable moment; the number of vesicles and the central positions of vesicles are determined from the set based on χ²-statistics of the residuals in least-square fits of the models to the image data. By comparing the detection accuracy with or without the “self-checking” algorithm in simulated images, we found that the detection accuracy with the “self-checking” algorithm was improved significantly. Meanwhile, we proposed an optimized flow chart of vesicle tracking which was applied to analyze the vesicles in mice β cells. We found that the number of vesicle traces will increase and the average docking time of vesicles will decrease after glucose stimulation based on our tracking analysis. This is because β cells will release insulin to regulate glucose balance with the help of vesicle translocation and secretion after glucose stimulation. In a word, we quantified the vesicles activity in mice β cell by tracking analysis on subcellular level.