Global and local feature interactive parallel multi-modal medical image fusion method

The invention discloses a global and local feature interactive parallel multi-modal medical image fusion method, and belongs to the technical field of medical image fusion. In order to solve the problems that in an existing multi-modal medical image fusion method, Transform is insufficient in local feature extraction capacity, and texture and edge information is lost in the feature extraction process, Local Transform block formed by a local multi-head attention mechanism is used in an Encoder module, and the local multi-head attention mechanism is used in the Encoder module. The problem of insufficient local feature extraction capability of Transform in a multi-modal medical image fusion method based on deep learning is solved; and a global and local feature interaction module is introduced in the feature extraction process, so that the loss of global and local information of the image is reduced. Experimental results show that the multi-modal medical fusion image obtained by the method provided by the invent