Enhancement Research Of Brain MRI Image Based On The Second Generation Of Curvelet Transform

The magnetic resonance imaging(MRI) with its no ionizing radiation, high contrastin soft tissue, multi parameter imaging and the advantages of good Three-dimensionalresolving power gradually become an important imaging method in medical imaging,especially in the study of brain pathology diagnosis.However,due to the paticularity ofthe MRI principle,its image will be more or less affected by heat and physiological noiseinterference in the process of acquisition and transformation and the brain singnals isweak in MRI,so the imaging results are not satisfactory. The details of the image can notbe observed,which bring many difficults to the doctor's diagnosis, so aiming to do someresearch about the method which can be used to effectively enhance the image of MRI fordoctor's diagnosis has extremely important clinical significance and application value.The common image enhancement methods contains histogram equalization method,median filtering, wavelet transform and so on,but among these methods,the source andconstituents elements of the image noise can't be analized thoroughly in the imageenhancement processing and the results is not ideal by using these methods,so thesemethods can't meet the requirement of the doctor's illness diagnosis.The noise will beenhanced by the tradtional ideas which make the weak edges and the noise difficult to bedistinguished causing the issue such as the poor contrast and the fuzzy effect during theenhancement processing. And the image details or edges will be weakened by thedenoising algorithm in noise filtering.In order to slove these problems,the enhancement algorithm of the brain MRI imageis proposed in this paper based on Second Generation of Curvelet Transform.The brainMRI image noise source and distribution characteristics are analized and the noise modelis established. And then the image is enhanced by the Second Generation of CurveletTransform Algorithm. The results shows that the processed image can not only filter thenoise,but also keep the details of the image,and at the same time the image contrast canbe improved and the goal of the brain MRI image enhancement is achieved.