Research On Fusion Algorithm Of Medical Images

With the continuous development of medical imaging technology, the type of medical image has become more complicated. Since single modal image has been unable to meet the requirement of clinical medicine, so it has become a hot research in medical images today how to use the effective information extracted from these complicated medical image data for the synthesis of the suitable clinical image. Medical image fusion technology is an effective tool to meet this requirement, this technology has already been widely applied in clinical medicine and provided an important reference for clinical diagnosis and treatment.In this paper, the research on medical image fusion algorithm is based on CT and MRI images. The main content of the paper include the following four parts:Firstly, the paper introduced the research background, significance and the present situation about medical image fusion technology. Then the paper also described and presented several traditional fusion algorithms, evaluated the fusion images in the subjective aspect, analyzed the advantages and disadvantages of each algorithm.Secondly, the paper introduced the wavelet transform and image fusion based on wavelet transform. For its shortcomings, the paper presented a new fusion rule based on regional features, improved fusion algorithm based on wavelet transform. Through the simulation experiments, the results showed that the algorithm is feasible and achieved better fusion effect.Once again, the paper introduced PCNN and used PCNN ignition map to get fused image. The paper still presented the link which based on the standard deviation, through the simulation experiments, the results showed that the algorithm is superior to other algorithms.Finally, the paper combined with the advantages of wavelet transform and PCNN to describe an improved fusion algorithm based on PCNN and wavelet transform, the low-frequency component based on regional features rule, the high-frequency component based on PCNN rule. Through the simulation experiments, the results showed that the improved algorithm is superior to single fusion algorithms based on PCNN or improved fusion algorithms based on wavelet transform.