An infrared image splicing method based on coarse-to-fine registration

Abstract

The invention discloses an infrared image splicing method based on coarse-to-fine registration, and the method comprises the steps: carrying out the coarse registration of an infrared image, carryingout the fine registration of the infrared image, improving the image registration precision, and improving the image splicing effect, wherein in the coarse registration process, feature points are extracted by utilizing every two adjacent images to construct a homography matrix, and each image can use an untransformed original image as an extraction object of the feature points in the coarse registration process, so that the extraction quantity of the feature points is ensured. The fine registration process is based on a fine registration sequence for sequential registration; deformation errors existing in the coarse registration process are adjusted, the feature point extraction process is obtained on the basis of the rule that the percentage of the area of the minimum circumscribed convex polygon covered by all feature point pairs consistent with the feature point pairs and the area of an original image is in direct proportion to the image registration accuracy, and the precision offine registration is guaranteed.

Description

technical field [0001] The invention belongs to the technical field of image stitching, and in particular relates to an infrared image stitching method based on coarse-to-fine registration. Background technique [0002] For image stitching technology, there are a lot of research on image stitching at home and abroad. Generally, image registration methods can be roughly divided into two categories: intensity-based and feature-based methods. Feature-based methods initially extract salient features and then use similarity metric to establish a geometric correspondence between two images. One of the main advantages of these methods is that they are fast and robust to noise, complex geometric distortions and significant radiometric differences. It performs well under the condition of matching suitable features. Commonly used features include points, edges, contours, and regions. Well-known feature matching methods include invariant descriptors, spatial relationships, and relaxati...

AI Technical Summary

Problems solved by technology

[0004] The existing infrared image registration method is generally to carry out image registration in sequence, which identifies the feature point pairs with the reference image, then calculates the homography matrix for registration, and then processes other images in turn for registration, and then the process Due to the accumulation of deformation errors, the final registration accuracy will not be high. On the other hand, the low contrast of infrared images leads to the lack of extraction of feature points, especially in the sequential registration method. The feature points that can be identified in the subsequent registration process are very small less, resulting in low accuracy of image registration and poor stitching image effect

Images