Algorithm Of Poisson-Noise Removing Based On ICA And Its Application On CT Imaging

Independent Component Analysis (ICA) which produced with Blind Source Separation is a novel signal processing method developed recently, it analyzes the data from a statistical point of view and becomes a powerful tool on analysis of multi-dimensional data. ICA aims at recovering the latent variables only from the observed mixtures. It is well known that, in the field of image analysis and pattern recognition, much of the important information of the image may be contained in the higher-order relationships among the image pixels. ICA based on higher-order statistics, has shown great promising ability in image feature extraction, voice recognition, medical signal processing. At present, ICA has become an important research aspect in Blind Source Separation and artificial neural network.This thesis presents the theories and algorithms of ICA and explores its applications in the image feature extraction, CT image denoising. The main work of the author focuses on the following aspects:1) First, give an overview of the status of ICA research.2) Introduce some ICA-related knowledge.3) Study the typical algorithms of ICA.4) Analyse the theory of CT imaging and use EGSnrc Monte Carlo simulation system to simulate CT imaging.5) In the experiments, we combine ICA with sparse coding to find an efficient coding of natural scenes. In addition, we use a thresholding operator on the coefficients to reduce the Poisson noise in CT images. In the process of denoising, we first use EGSnrc Monte Carlo simulation system to rebuild the source image.We also compare the ICA method with wiener filter algorithm. The comparisons show ICA has better performance in image feature extraction and CT image denoising with simulated data.