Image matching method and device and computer readable storage medium

Abstract

The invention relates to the technical field of artificial intelligence, and discloses an image matching method, which comprises the following steps of generating an image imaging picture according toa scene image shot by an aerial photography instrument, carrying out primary image matching on the image imaging picture by utilizing a scale invariant feature transformation method, and generating aprimary image matching set; based on the primary image matching set, generating an epipolar line image, calculating the overlapping degree between the epipolar line images, completing secondary imagematching, and generating a secondary image matching set; and based on the second image matching set, establishing dense matching of all pixel points between images, generating a third image matchingset, and executing three-dimensional reconstruction to obtain a reconstructed scene image. The invention further provides an image matching device and a computer readable storage medium. The inventionprovides a novel image matching scheme applied to a dense three-dimensional scene under aerial photography, and the image matching efficiency can be improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to an image matching method, device and computer-readable storage medium. Background technique [0002] Image matching refers to the process of identifying points with the same name between two or more images through a certain matching algorithm. It is an important early step in image fusion, target recognition, target change detection, computer vision, etc. It has a wide range of applications in remote sensing, digital photogrammetry, computer vision, cartography, and military applications. At present, the most effective and effective method of image matching is to judge the difference of the image through visual inspection; secondly, according to the principle that the essence of the image is the pixel, compare the gray value of all pixels in the target area; or find the target image based on the principle of template matching The same or most similar position as the su...

AI Technical Summary

Problems solved by technology

Due to the large number of aerial images and the dense objects in each image, it is impractical for human eyes to make visual inspections; and when using pixel matching to make differences between multiple image pixels, it is subject to noise, quantization errors, tiny Illumination changes, minimal translation and other effects will produce large pixel differences and affect the matching effect; while the template matching method is applied to dense images, it needs to generate a large number of matching modules to search in the image, and the timeliness is average. , and affected by image noise, the probability of mismatching is also very high

In general, due to the complexity and changeability of dense scenes under aerial photography, 3D reconstruction can help users conduct analysis and research more effectively, and the above methods lack this function

Images