# US 12,002,260 B2 · Automatic topology mapping processing method and system based on omnidirectional image information > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,002,260 B2 - Document ID: US12002260B2 - Status: Granted - Kind: B2 - Application number: 17/429,616 - Filed: 2020-06-22 - Published: 2022-05-05 - Granted: 2024-06-04 - Earliest claimed: 2019-06-21 - Assignees: 3I Inc. - Applicants: 3I Inc. - Inventors: Gyu Hyon Kim; Farkhod Khudayberganov - Web page: https://patentgrep.com/patent/US12002260B2 ## CPC - G06V 10/82: using neural networks - G06V 20/70: Labelling scene content, e.g. deriving syntactic or semantic representations - G06T 7/32: using correlation-based methods - G06V 10/16: {using multiple overlapping images; Image stitching} - G06V 10/25: Determination of region of interest [ROI] or a volume of interest [VOI] - G06V 10/469: {Contour-based spatial representations, e.g. vector-coding} - G06V 10/757: {Matching configurations of points or features} - G06V 10/7715: {Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods} - G06V 20/36: {Indoor scenes} - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract An automatic topology mapping processing method and system. The automatic topology mapping processing method includes the steps of: obtaining, by the automatic topology mapping processing system, a plurality of images, wherein at least two of the plurality of images include a common area in which a common space is captured; extracting, by the automatic topology mapping processing system, from respective images, features of the respective images through a feature extractor using a neural network; and determining, by the automatic topology mapping processing system, mapping images of the respective images on the basis of the features extracted from the respective images. ## Claims ### Claim 1 An automatic topology mapping processing method comprising: acquiring, by an automatic topology mapping processing system, a plurality of images, wherein at least two of the plurality of images include a common area in which a common space is captured; extracting, by the automatic topology mapping processing system, features of the images from each of the images through a feature extractor using a neural network; and determining, by the automatic topology mapping processing system, mapping images of the images based on the features extracted from each of the images; wherein the determining, by the automatic topology mapping processing system, mapping images of the images based on the features extracted from each of the images comprises: constructing a database (DB) containing vectors representing the features extracted from each of the images; performing a vector search using a vector set that includes at least some of first vectors corresponding to first features extracted from a predetermined first image among the images from the constructed DB, wherein only the features corresponding to a predefined area in the image are input into the vector search to identify a positional relation of the plurality of images, and acquiring a closest vector set to the vector set; and determining a second image corresponding to the closest vector set as a mapping image of the first image. ### Claim 2 (depends on 1) The method according to claim 1, further comprising mapping the first image and the second image determined as a mapping image of the first image. ### Claim 3 (depends on 2) The method according to claim 2, wherein the mapping the first image and the second image determined as a mapping image of the first image comprises: determining first feature-corresponding positions on the first image corresponding to the first features extracted from the first image, respectively, and second feature-corresponding positions on the second image corresponding to second features extracted from the second image, respectively; and determining a relative positional relation between the first image and the second image based on the determined first feature-corresponding positions and the second feature-corresponding positions. ### Claim 4 (depends on 1) The method according to claim 1, wherein the neural network is a network trained to output a transformation relation so that points corresponding to each other extracted from an overlapping common area of divided images divided from a predetermined image to have an overlapping area optimally match. ### Claim 5 (depends on 1) The method according to claim 1, wherein the images are 360-degree images photographed at different positions in an indoor space. ### Claim 6 (depends on 1) A non-transitory computer readable medium having stored thereon software instructions that, when executed by a processor, cause the processor to generate control signals to perform the method according to claim 1. ### Claim 7 An automatic topology mapping processing system comprising: a processor; and a memory for storing a program implemented by the processor, wherein: the program acquires a plurality of images; and at least two of the plurality of images include a common area in which a common space is captured, extracts features of the images from each of the images through a feature extractor using a neural network, and determines mapping images of the images based on the features extracted from each of the images, wherein the program constructs a database (DB) containing vectors corresponding to the features extracted from each of the images, performs a vector search using a vector set that includes at least some of first vectors corresponding to first features extracted from a predetermined first image among the images from the constructed DB, wherein only the features corresponding to a predefined area in the image are input into the vector search to identify a positional relation of the plurality of images, and acquiring a closest vector set closest to the vector set; and determines a second image corresponding to the closest vector set as a mapping image of the first image. ### Claim 8 (depends on 7) The system according to claim 7, wherein the program constructs the database (DB) containing vectors corresponding to the features extracted from each of the images, performs the vector search using a vector set that is at least some of first vectors corresponding to first features extracted from the predetermined first image among the images from the constructed DB, and determines the second image extracted based on a result of the vector search as a mapping image of the first image. ### Claim 9 (depends on 8) The system according to claim 8, wherein the program determines first feature-corresponding positions on the first image corresponding to the first features extracted from the first image, respectively, and second feature-corresponding positions on the second image corresponding to second features extracted from the second image, respectively, and determines a relative positional relation between the first image and the second image based on the determined first feature-corresponding positions and the second feature-corresponding positions.