This paper is devoted to noise suppression in systems for double station observation of meteors, nowadays known as MAIA (Meteor Automatic Imager and Analyzer). The noise analysis based on acquisition of testing video sequences at different lighting conditions and their statistical evaluation were described in our previous paper. The measurement showed that the type of noise generated by the system is signal-independent in a certain illumination range. The noise and image models in the wavelet domain are based on the Generalized Laplacian Model (GLM) and it is the most convenient to estimate the model parameters using the moment method. Furthermore, the noise component may be modeled by the GLM also in the space domain. Overall, we verified that the GLM allows for modeling various types of probability density functions. In the final section of this paper, the performance of the proposed advanced de-noising algorithm is verified on the real data, which were acquired in the Astronomical Institute.