A method, system, computer device and medium for calibrating a continuous panorama

By employing feature matching and 3D rotation matrix optimization methods, the problem of noise interference in panoramic image calibration due to feature matching is solved, achieving high-precision and robust calibration of panoramic images, which is suitable for automatic calibration in complex scenarios.

CN122115195APending Publication Date: 2026-05-29GUANGZHOU MASHI INFORMATION TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
GUANGZHOU MASHI INFORMATION TECH CO LTD
Filing Date
2026-02-03
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing panoramic image calibration techniques are susceptible to feature matching noise in complex scenes, making it difficult to achieve automatic calibration using only two panoramic images. In particular, the calibration accuracy and stability are insufficient under conditions of significant displacement.

Method used

Feature points of the panoramic image are extracted using a feature matching algorithm, converted into three-dimensional spherical coordinates, and the three-dimensional rotation matrix is ​​optimized using a random sampling consistency algorithm. Image reprojection is then performed to achieve orientation consistency calibration of the panoramic image.

Benefits of technology

It achieves high-precision and robust calibration of panoramic images in complex environments, improving the accuracy and adaptability of calibration and avoiding image distortion and mismatch problems in traditional methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122115195A_ABST
    Figure CN122115195A_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of digital image processing, and particularly relates to a continuous panoramic image calibration method and system, computer equipment and medium; the method comprises the following steps: acquiring a first panoramic image and a second panoramic image photographed at different spatial positions; performing feature point extraction and matching on the first panoramic image and the second panoramic image to obtain a plurality of feature point matching pairs; converting image coordinates in the feature point matching pairs into corresponding three-dimensional spherical coordinates; calculating a three-dimensional rotation matrix based on the three-dimensional spherical coordinates; obtaining an optimized rotation matrix through multiple random sampling and inlier screening, and projecting the second panoramic image onto a spherical surface according to the optimized rotation matrix to perform a rotation transformation process on the second panoramic image, so that the shooting direction of the second panoramic image is consistent with that of the first panoramic image. In this way, automatic calibration can be completed only by relying on the image features of two panoramic images, and the accuracy and environmental adaptability of panoramic image direction calibration are improved.
Need to check novelty before this filing date? Find Prior Art