Medical image fusion method based on multi-scale mechanism and residual attention and medium

Abstract

The invention discloses a medical image fusion method based on a multi-scale mechanism and residual attention and a medium, and the method comprises the steps: S1, inputting an anatomical image and a functional image after registration into a convolution kernel with the size of 1 * 1, and increasing the dimensions of input features; S2, respectively inputting the registered anatomical image and functional image into a multi-scale mechanism of two branches, extracting feature maps of the anatomical image and the functional image on different scales, then inputting the extracted feature maps into a residual attention network, and further extracting features of the input images; S3, fusing the extracted feature maps of the anatomical image and the functional image; and S4, reconstructing the fused feature map through three-layer convolution to obtain a final fused image. The problems of information loss, color distortion and the like when the pseudo-color image and the grayscale image are fused in a medical image fusion method are effectively solved.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a medical image fusion method based on multi-scale residual attention. Background technique [0002] Medical image fusion methods belong to the field of computer vision, and have a wide range of applications in medical imaging and clinical diagnosis. In order to eliminate the limitations of single-modal images in information expression, multi-modal medical images are fused through different fusion algorithms. Multi-scale transformation is an approximation and simulation of human vision. Commonly used multi-scale transformation algorithms include pyramid, wavelet, non-subsampled contourlet, and non-subsampled shearlet. [0003] The image fusion method based on the Laplacian pyramid has salient features of different scales and different resolutions, and can obtain fusion effects close to human visual characteristics. However, pyramid-based methods have block effects and...

AI Technical Summary

Problems solved by technology

[0004] Although there are many fusion methods, there are still many challenges in the actual scene, such as the influence of noise, the quality of the image to be fused, color distortion, image artifacts, etc.

Although the current convolutional neural network-based fusion method has greatly improved the preservation of texture and color information, the fusion image obtained by the convolutional neural network-based algorithm is usually too smooth

Images