Method and system for real-time splicing of microscopic images and computer equipment

Abstract

The invention discloses a method and system for real-time splicing of microscopic images and computer equipment. According to the method, a two-stage dual-network cascaded microscopic image splicing model is adopted; a pixel-level image registration network is used to directly output a pixel-level mapping table from an input image pair, and therefore, pixel-by-pixel mapping is realized, and each pixel can be aligned to an ideal alignment position. Compared with a common method of calculating a homography matrix or calculating homography matrixes corresponding to a plurality of grids, the method of the invention, based on the pixel-level mapping table, is high in precision, high in speed and capable of performing real-time processing; and a fused image is directly output from the input aligned images by using a multi-focus image fusion network, so that multi-focus image fusion is realized, all parts of the obtained spliced image are completely clear, and accurate splicing can also be realized under the condition that a weak texture region exists in an image pair overlapping region.

Description

technical field [0001] The invention relates to the technical field of microscopic image processing, in particular to a method, system and computer equipment for real-time splicing of microscopic images. Background technique [0002] With the development of digital microscope technology, the existing technology can use the sample slide scanning method to scan the sample slide comprehensively to obtain a full-section digital microscopic image, that is, to sequentially take pictures of each partial field of view of the sample slide under the magnification of the digital microscope Microscopic images, and use image stitching technology to obtain a complete digital microscopic image of the whole section. [0003] Generally speaking, the mosaic technology of microscopic images includes two key steps, which are image registration and image fusion. Currently, the widely used image mosaic technology is based on feature point matching. Feature extraction is performed on overlapping ...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a method, system and computer equipment for real-time splicing of microscopic images, which are used to overcome defects in the prior art that do not meet real-time performance, and may cause image splicing failures even when there are weak texture areas in overlapping areas.

Images