Abstract: A theoretical model for Kalman filtering algorithm is established based on the imaging characteristic of the confocal microscope. The Kalmam filtering algorithm based on the image pixel is proposed and a realtime Kalman filter is realized by means of the introduction of Kalman filtering method into the system. The experimental result shows that the algorithm can effectively improve the signal-noise-ratio (SNR) for the confocal microscopy imaging systems, but it is obtained at the expense of time. Therefore, the perfect method to improve the resolution of the system is to emphatically consider the optimization of the beam path of the imaging system and the detection circuit.