In some cases the number of projections in a set of tomography data is limited. This can be seen from their sinogram. If this kind of data is reconstructed, the image produced will be deteroited. In order to overcome this problem an algorithm for restoring the incomplete sinogram has been developed in previous work. In that work the developed algorithm was tested using ideal data from Shepp-Logan panthom. In this paper the same algorithm is applied to restore incomplete and noisy data in ultrasound tomography. The phantom used in this study is a concrete cylinder with a diameter of 10.62 cm. Inside cylinder there are 6 holes with a diameter of 2 cm each. These holes were filled with various materials. The phantom were scanned in ultrasound tomography scheme and the resulted sinogram then reconstructed with and without sinogram restoration. Using contrast to noise ratio (<em>CNR</em>) indicator the reconstructed image with sinogram restoration is better than that without sinogram restoration.
Sinogram restoration based on shape property in computed tomography