Accurate attenuation correction of emission data is mandatory for quantitative analysis of PET images. One of the main concerns in CT-based attenuation correction (CTAC) of PET data in multimodality PET/CT imaging is misalignment occurred due to respiratory artifact between PET and CT images. In this paper a combined method which is simple and fast is proposed for registration of PET and CT data to correct the effect of this artifact. The algorithm is composed of two step: First step is meant to reduce the noise by applying an adaptive gradient anistropic diffusion filter then using Iterative closest point (ICP) registration method in order to obtain initial estimation to ensure fast and accurate convergence of the algorithm. At the second step, the respiratory related artifact of PET images is greatly reduced by employing Free Form Deformation algorithm based on B-spline which provides more accurate adaptive transformation to align the images