Training method, apparatus and image display system for image stitching

Abstract

The present invention relates to an image processing system configured to perform a computer-implemented method for generating stitched images using a convolutional neural network, a computer-implemented method for generating stitched images using a convolutional neural network, and a method for training a convolutional neural network to determine image representations for image stitching, the training of the convolutional neural network comprising the steps of receiving image data, the image data having at least two images overlapping each other at an overlap region where the images overlap each other in a target anatomical structure, the overlap region having an optimal displacement between the two images such that the two images can be correctly combined for image stitching, determining a cost function of the displacement of the two images in the overlap region, determining a deviation of the cost function from a reference cost function, and optimizing the convolutional neural network based on the deviation.

Images