Journal of Applied Sciences

• Articles • Previous Articles     Next Articles

Image Reconstruction from Limited Range Projections Using Discrete
Orthogonal Moments

DAI Xiu-bin 1;4, SHU Hua-zhong 1;4, LUO Li-min 1;4, Han Guo-niu 2, Jean Louis Coatrieux 3;4   

  1. 1.Laboratory of Image Science and Technology, Southeast University, Nanjing 210096, China; 2. Universite Louis Pasteur et C.N.R.S., Strasbourg 67084, France; 3. Universite de Rennes 1, Rennes 35000, France; 4. Centre de Recherche en Information Biomedicale Sino-Francais, Nanjing 210096, China
  • Received:2008-02-22 Revised:2008-07-19 Online:2008-12-10 Published:2008-12-10
  • Contact: DAI Xiu-bin

Abstract: Estimation of projection from unknown views is an efficient way to improve image reconstruction quality when parts of the projections are missing. This paper presents a new method to estimate unknown projections. To achieve better results, we make use of the discrete Radon transform and a set of discrete orthogonal Tchebichef polynomials to establish relationship between the projection moments and the image moments, and then estimate the unknown projections from the known projections using this relationship. Experimental results show that the proposed method provides good image reconstruction quality, and is robust against noise.

Key words: discrete Radon transform, Tchebichef orthogonal moments, limited-view CT image reconstruction

CLC Number: