A Medical Image Stitching Method Based on Mutual Information and Interaction

Abstract

The invention provides a mutual information and interaction-based medical image splicing method. According to the method, a strategy for searching for an image registration parameter is improved in a way that a target area and a comparison area are simultaneously enlarged, images are sequentially registered in a horizontal direction and a vertical direction, and are fused and seamlessly processed according to the obtained registration parameter, a linear weighting function is used for smoothly transiting an overlapped area, and a strategy for determining the size of the area of a spliced result image is also improved to realize the automatic and seamless splicing of images in any number to finally obtain an extra large-viewing angle medical image. In addition, in order to increase splicing speed and meet the requirements of a doctor, interaction with the doctor is added when the method is implemented, the doctor can manually divide an area to be matched, and a splicing process can be simplified according to needs. Experiments prove that a satisfactory splicing result can be obtained by the method and that the method has high clinical application value and is significant for researches.

Description

technical field [0001] The invention relates to the fields of computer vision and digital image processing, in particular to a new method for medical image stitching. Background technique [0002] In recent years, medical images have become one of the fastest growing fields in medical technology, especially the development of computer technology and the emergence of new imaging technologies and equipment such as X-ray, MRI, CT, PET, etc., making medical image processing technology more important for medical scientific research. and the growing role and influence of clinical practice. However, due to the mechanical limitations of medical imaging equipment, generally only partial images can be obtained, and panoramic images cannot be obtained. When a doctor wants to observe the comprehensive information of multiple images, he usually can only spatially align several images through his own observation. This method is often highly subjective, with low reliability and poor repea...

AI Technical Summary

Problems solved by technology

However, due to the mechanical limitations of medical imaging equipment, generally only partial images can be obtained, and panoramic images cannot be obtained

When a doctor wants to observe the comprehensive information of multiple images, he usually can only spatially align several images through his own observation. This method is often highly subjective, with low reliability and poor repeatability. large error

Images