# US 11,443,443 B2 · Method and a data processing system for aligning a first panoramic image and a second panoramic image in a navigation procedure > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,443,443 B2 - Document ID: US11443443B2 - Status: Granted - Kind: B2 - Application number: 16/930,888 - Filed: 2020-07-16 - Published: 2022-01-20 - Granted: 2022-09-13 - Earliest claimed: 2020-07-16 - Assignees: Siemens Industry Software Inc. - Applicants: Siemens Industry Software Inc. - Inventors: Vladislav Murashkin - Web page: https://patentgrep.com/patent/US11443443B2 ## CPC - G06T 7/33: using feature-based methods - G06T 3/4038: Image mosaicing, e.g. composing plane images from plane sub-images - G06T 7/30: Determination of transform parameters for the alignment of images, i.e. image registration - G06T 7/38: Registration of image sequences - G06V 10/443: {by matching or filtering} - G06T 2207/30252: Vehicle exterior; Vicinity of vehicle - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A method aligns a first panoramic image and a second panoramic image in a navigation procedure. The method includes using a computer vision algorithm for extracting features that are present in the panoramic images and use the extracted features to prepare a first feature vector for the first panoramic image and a second feature vector for the second panoramic image. The first and the second feature vectors are compared and matches in both the first and the second feature vectors that are above a determined threshold are identified. For the identified matches in each of the first and the second feature vectors for each of the two images, the perspective angle between a reference point of the respective image and the features according to the identified matches are determined. Finally, a difference between the determined perspective angles for a part of the identified matches in the panoramic images is determined. ## Claims ### Claim 1 A method for aligning a first panoramic image and a second panoramic image in a navigation procedure, which comprises the following steps of: using a computer vision algorithm for extracting features that are present in the first panoramic image and/or in the second panoramic image and using extracted features to prepare a first feature vector for the first panoramic image and a second feature vector for the second panoramic image; comparing the first and the second feature vector and identifying matches in both the first and the second feature vector that are above a determined threshold; determining for identified matches in each of the first and the second feature vectors for each of the first and second panoramic images a perspective angle between a reference point of a respective one of the first and second panoramic images and the features according to the identified matches; and determining a difference between determined perspective angles for at least part of the identified matches in the first and the second panoramic image. ### Claim 2 (depends on 1) The method according to claim 1, wherein the reference point is chosen to be a center of a respective one of the first and second panoramic images. ### Claim 3 (depends on 1) The method according to claim 1, wherein for at least part of the identified matches a center of a respective feature is identified and the determined perspective angles are determined based on an identified center and the reference point. ### Claim 4 (depends on 1) The method according to claim 1, wherein the difference between the determined perspective angles forms an array of angle differences. ### Claim 5 (depends on 4) The method according to claim 4, wherein a trimmed mean of values for an angle difference in said array are considered to be a rotation difference between the first and second panoramic images. ### Claim 6 (depends on 3) The method according to claim 3, wherein the identified center is a center of a respective one of the first and second panoramic images. ### Claim 7 (depends on 4) The method according to claim 4, wherein the rotation difference is applied in the navigation procedure for a spatial alignment of the first and second panoramic images. ### Claim 8 A data processing system, comprising: a processor; an accessible memory coupled to said processor; and the data processing system being configured to perform an alignment procedure to align a first panoramic image and a second panoramic image in a navigation procedure, said alignment procedure containing the following steps of: using a computer vision algorithm for extracting features that are present in the first panoramic image and/or in the second panoramic image and using extracted features to generate a first feature vector for the first panoramic image and a second feature vector for the second panoramic image; comparing the first and the second feature vector and identifying matches in both the first and the second feature vector that are above a determined threshold; determining for identified matches in each of the first and the second feature vector for each of the first and second panoramic images a perspective angle between a reference point of a respective one of the first and second panoramic images and the features according to the identified matches; and determining a difference between determined perspective angles for at least part of the identified matches in the first and the second panoramic image. ### Claim 9 (depends on 8) The data processing system according to claim 8, wherein the reference point is chosen to be a center of a respective one of the first or the second panoramic image. ### Claim 10 (depends on 8) The data processing system according to claim 8, wherein for at least part of the identified matches a center of the respective feature is identified and the perspective angles are determined based on an identified center and the reference point. ### Claim 11 (depends on 8) The data processing system according to claim 8, wherein a difference between the determined perspective angles forms an array of angle differences. ### Claim 12 (depends on 9) The data processing system according to claim 9, wherein a trimmed mean of values for an angle difference in said array are considered to be a rotation difference between the first and second panoramic images. ### Claim 13 (depends on 10) The method according to claim 10, wherein the identified center is a center of a respective one of the first and second panoramic images. ### Claim 14 (depends on 12) The method according to claim 12, wherein the rotation difference is applied in the navigation procedure for a spatial alignment of the first and second panoramic images.