Lens defocus causes image blurring. Restoring these kind of images is an interesting research field. The first and most important step of restoring images is blur identification. Since now, many researchers have presented methods to estimate out of focus blur and restore original images, but most of these methods suffer the additive noise. In this paper we have presented a method to estimate out of focus blur precisely, such that it does not depend on the noise level. This method uses two rough estimates of blur function parameters, then it calculates the real parameter using an iterative algorithm. Wiener filter, Bessel function and some mathematical relations properties are the bases of this method. Experimental results showed our method gives high precision.