Radar imagery is nowadays used to observe ocean waves despite the fact that radar images contain invisible areas because of the shadowing effect in the radar mechanism. Moreover, the radar images show the radar intensity which is not directly related to the wave height. This paper deals with the subject to estimate the significant wave height $H_s$ from (synthetic) radar images and will show that the change of shadowing with distance from the radar is the key property to estimate $H_s$. In fact, the extent of shadowing, quantified by a visibility function, depends on various physical variables, but most notably on two dimensionless quantities: the ratio of radar height and $H_s$ in the vertical direction, and the normalized distance in the horizontal direction. Assuming the normalized sea spectrum to be known, or approximating the spectrum, visibility functions can be determined for various $H_s$, defining the ingredients of a database. Comparing an observed visibility from successive images with the elements of the database, a best-fit approximation will provide an estimate of the $H_s$ of the observed sea. Randomness of the sea will only slightly affect the observed visibility and Monte Carlo simulations will annihilate these effects in the database.