Various image processing algorithms for acquiring and tracking an inter-satellite laser beam are assessed. These algorithms aim at detecting a faint point target of a few picowatts embedded in a background by means of acquisition systems based on CCD sensors. Simple algorithms for real-time operation have been designed. A model which takes into account the background, its spatial and temporal fluctuations as well as the opto-electrical parameters of the CCD has been developed. Fluctuation of the background and CCD noise appear to be the main perturbations that hinder obtaining the target's spatial coordinates. Two types of image processing techniques have been analyzed: single frame processing using spatial filtering and multiple frame processing using frame subtraction. The performed analyses show that in systems where fixed spatial fluctuations are the main constraints, the frame subtraction method is the most efficient in terms of probability of acquisition. Whereas the spatial filtering method is much simpler to implement, it can offer a significant improvement if the spectral energy of the fixed pattern parasitic background is low within the passband of the chosen spatial filter. The performances of these techniques are compared in terms of successful target acquisition.