Self-adaptation splicing method and system of X-ray images

Abstract

The invention discloses a self-adaptation splicing method of X-ray images. The method comprises the following steps that 1. horizontal image bands of two continuous X-ray images are obtained; 2. the variances of the two horizontal image bands are computed respectively, and a template image and a moving image are determined; 3. the template image and the moving image are registered; and 4. an overlap region nonlinear fusion method is used for carrying out fusion on images, and a new image after splicing is obtained. According to the method, the variances of the gray values of the horizontal image bands selected from the overlap region part of the two X-ray images are compared respectively, a clear horizontal image band with more information is selected in a self-adaptation mode as a template to be registered with another image, then the overlap region nonlinear fusion transition method is used for carrying out fusion on input images, and full-scale splicing is achieved.

Description

technical field [0001] The invention belongs to the technical field of digital image retrieval, and in particular relates to an X-ray image self-adaptive mosaic method and a mosaic system thereof. The invention is used for automatically splicing X-ray images in different regions collected by X-ray imaging equipment. Background technique [0002] With the advent of the digital age, the digitization of X-ray images has become a trend. X-ray images greatly facilitate the examination and location of the patient's lesions by doctors, and play a great role in the prognosis and prognosis of the disease. At present, digital image processing algorithms such as image enhancement, edge detection, contour extraction, image registration and stitching, image coding, etc. have been widely used, bringing great convenience to doctors and reducing the rate of misdiagnosis. Existing X-ray imaging equipment (such as CR, DR, etc.) has a limited image size (approximately 17×17 inches) at a time....

AI Technical Summary

Problems solved by technology

First of all, image registration algorithms are mainly divided into feature-based registration algorithms and gray-scale information-based registration algorithms. Since X-ray images generally have fewer feature points, feature-based image registration methods are often less accurate than medical images. Therefore, the current research hotspots of medical image registration mainly focus on the registration algorithm based on gray information, and the registration algorithm based on normalized mutual information in this algorithm has better robustness. However, the traditional normalized mutual information method image registration algorithm fixedly selects the overlapping area of ​​one of the images as a template image for registration with another image when registering images. However, the image information contained in the overlapping area of ​​the two images is different, which leads to a large change in the final result with the template selection.

Secondly, common image fusion methods include average value method and overlapping area linear transition method, among which the more commonly used method is overlapping area linear transition method, which uses linear transformation weights to fuse images to make the transition area look softer, but, This method cannot fully eliminate the effects of overexposure on the edge of the two images to be registered, excessive deformation of the edge of the image, etc. on the accuracy of image registration.

Images