Skip to Main Content
In most cases, ill-posedness exists inherently in the solution process of inverse image problems (such as X-ray CT image reconstruction from projections, ECG and EEG inverse mapping), and affects seriously the quality, stability and accuracy of the reconstructed images with the susceptibility to the errors in measurement data and numerical solution of forward problems. This paper gave a study on the ill-posedness of image reconstruction from projections and discuss the characteristics of the ghost function which affects imaging quality and accuracy in the reconstruction process. Multicriterion regularization approach, a new approach to solve the ill-posed inverse problems, was presented with its theory basis and the experiment results. The solution stability and accuracy were studied using singular value decomposition (SVD) theory, and main factors affecting the reconstruction quality are discussed.