Research On Medical Image Fusion Based On Wavelet Transform

In recent years, image fusion has become an important and useful technique for image analysis and computer vision, especially in medical domain. Due to the difference between imaging mechanism of medical instrument and application environment, the medical images obtained have different details, using the complementary or redundancy information of several sources medical images, we can make the fusion image contain more abundant details and more complete information, the fused image is more complete and accurate than the single source image.In this dissertation, we focus on the problem of image fusion. The three levels of image fusion including pixel-level fusion, feature-level fusion and decision-level fusion are discussed in detail. The simple image fusion method, the image fusion techniques based on multi-scale pyramid decomposition, and the image fusion method based on wavelet transform are studied respectively. This dissertation is mainly focus on image fusion based on wavelet transform, for the image transformed, after considering several methods such as max-value selection, area energy fusion and area square difference fusion, we bring forward other fusion methods of max-value of frequency domain, .consistency in frequency domain and fuzzy fusion. Theimage fusion methods using biorthogonal wavelet transform are given, these fusion approaches have been successfully used in image fusion.Another main contribution of the thesis is the objective evaluation of image fusion performance. Quantitatively assessment of the image fusion performance is an important and complicated issue. Several evaluation criteria of information entropy, across entropy, PSNR, average square difference and definition are presented in this dissertation. Using these evaluation criteria, we study that different fusion rules influence the image fusion performance. These evaluation criteria are also applied to compare the performance of various image fusion methods.A series of experiments on image fusion are given in this dissertation. The experimental and calculation results show that the fusion schemes and evaluation criteria are effectual.