Research On Medical Image Fusion Algorithm Based On Improved PCNN Model

Medical image fusion is an intersection of the disciplines of integrated medical imaging,digital image processing,digital modelling and computer vision.With the maturity of computer technology,the application of multimodal medical image fusion technology is positively contributing to image quality assurance,which is of great significance in assisting medical diagnosis,fine surgical planning and scientific research progress.At present,research on medical image fusion algorithms is gradually becoming the focus,while certain results have been achieved for the algorithms,however,the current fusion algorithms still have shortcomings.This paper focuses on two aspects of the fusion process based on multi-scale decomposition of images and the use of impulse-coupled neural networks as fusion rules,and the following improvement work has been carried out to address the problems that exist in these two aspects.(1)To address the complex problem of setting PCNN parameters,meanwhile,the EOA algorithm has the advantages of fewer parameters,less time consuming,as well as strong optimisation finding ability,this paper proposes a medical image fusion algorithm based on NSST and EOA to optimise PCNN.First,the source image is decomposed using the NSST transform.Then,the three parameters of connection coefficient?,threshold attenuation coefficient?_Land threshold amplification coefficient V_?of the PCNN model are optimised by the EOA algorithm,and the above-mentioned optimisation method and region energy maximisation are used as the fusion rules for the low-frequency and high-frequency sub-bands respectively.Finally,the fused image is obtained by the NSST inversion and the fused image is evaluated by five image fusion evaluation metrics,Q^AB/F,AVG,MI,SF and SD.(2)To address the problem that multi-scale decomposition methods can only obtain a single feature type of an image when decomposing an image,while Maclaurin expansion decomposes the source image to a certain extent to obtain the complementary features of the image and also compensate for the above-mentioned defects in image decomposition,however,there is still the problem that the decomposed image has a low grey value and retains less energy information,this paper proposes a medical image fusion algorithm based on Maclaurin expansion and This paper proposes a medical image fusion algorithm based on Maclaurin expansion and PCNN.Firstly,the source image is decomposed using Maclaurin expansion to obtain the deviated component image and the multi-level energy component image,and the decomposed three-level energy component image is optimised by using Gaussian homomorphic filtering.Secondly,based on the algorithm proposed in Chapter 3,the improved Laplace energy and the external excitation of the optimised PCNN are used as the fusion rules for the deviance components and the region energy algorithm is used as the fusion rule for the energy components.Finally,the fused images are obtained by inverse Maclaurin.(3)A brain medical image fusion system was designed which integrates the algorithm proposed in this paper with other algorithms for the convenience of other users.