Research On Multimodal Medical Image Fusion Algorithm Based On Two-scale Decomposition

Multi-modal medical image fusion is the important branches of image fusion,which refers to the process of fusing two or more medical images from different imaging modalities into one image.Compared with single-modal medical images,the fused image contains richer information,clearer texture structure,and more obvious lesions.It can improve the accuracy and efficiency for doctors in clinical diagnosis and treatment.At present,multi-modal medical image fusion mainly focuses on the fusion of anatomical images(CT/MRI,MRI-T1/MRA,etc.)and the fusion of anatomical images with functional images(MRI/PET,MRI/SPECT,etc.).This paper also studies the fusion of these two kinds of medical images.For the problems of lower efficiency or poor fusion quality of current mainstream multimodal medical image fusion methods,two effective multimodal medical image fusion methods are proposed in this paper.The main innovations of this paper are as follows:1)To achieve an optimal balance between computational loss and fusion quality,this paper proposes a new multimodal medical image fusion method based on weighted local energy matching degree and improved spatial frequency(SF).Firstly,the latent low-rank representation is employed to learn a decomposition matrix that is applied to decompose a source image into base part and saliency part.Then,in order to adaptively select a fusion strategy for the base part,this paper proposes a novel fusion rule,namely weighted local energy matching measurement,which is constructed by calculating the matching degree between the corresponding pixels of the base parts.Meanwhile,after considering the characteristics of the main diagonal SF and the secondary diagonal SF,a fusion strategy based on improved SF and L2 norm is designed to merge saliency parts.Finally,the fused image is obtained by combining the fused base part and saliency part.Various experiments on multiple groups of medical images demonstrate that compared with the state-of-the-art fusion methods,the proposed method has superior performance in subjective visual and objective index evaluations.2)Based on the above problems,combined with the current medical image needs to be observed at a higher resolution,a multimodal medical image super-resolution fusion method based on edge detection and visual saliency mapping is proposed.Firstly,the source image is decomposed into base layer and detail layer by the method of two scale decomposition based on gradient operator.Then,the base layer is fused by weighted local energy to preserve the energy information of the source image in the fusion image as much as possible.For detail layer,in order to extract the details of high frequency images as much as possible and keep the significant feature information of the image,a fusion rule of detail layer based on edge detection and visual saliency mapping is designed.Finally,the fused base layer and detail layer are combined to get the fused image.In addition,this paper put forward for the first time to introduce medical image super-resolution into the field of multi-modal medical image fusion.The experimental results on multiple group medical images show that the proposed method is superior to the existing multimodal medical image fusion methods on subjective visual effect and objective indexes.Secondly,the proposed super-resolution of medical images can effectively improve the fusion quality of multimodal medical image fusion results.In particular,the proposed method is also suitable for multimodal medical image fusion with different resolution.