APPARATUS AND METHOD OF DENOISING A NOISE CHARACTERISTICS IN RECONSTRUCTED IMAGE

摘要

The present invention provides an apparatus and a method for removing a noise from a reconstructed image, which remove a Gaussian-based noise in consideration of a noise characteristic of a reconstructed image obtained in a medical diagnosis imaging apparatus, such as a positron emission tomography (PET), an SPECT, a CT, or a panorama X-ray. According to the present invention, the apparatus for removing the noise from the reconstructed image obtained in the medical diagnosis imaging apparatus comprises: a sensitivity map generation unit which generates a sensitivity map according to a system matrix corresponding to the medical diagnosis imaging apparatus; an average/dispersion linear unit which linearly makes a relationship between an average and a dispersion of an 3D image reconstructed in the medical diagnosis imaging apparatus by using the sensitivity map generated by the sensitivity map generation unit; a dispersion stabilization conversion unit which converts the reconstructed 3D image, to which the sensitivity map is applied, into Gaussian distribution; and a Gaussian noise removal unit which removes a Gaussian-based noise with respect to the reconstructed 3D image having the Gaussian distribution by the dispersion stabilization conversion unit.