System and method for determining the geographic location in an image

Patent metadata

Field	Value
Publication number	US-12579772-B2
Application number	US-202217943616-A
Country	US
Kind code	B2
Filing date	Sep 13, 2022
Priority date	Jan 5, 2021
Publication date	Mar 17, 2026
Grant date	Mar 17, 2026

Abstract

Methods and media for determining a list of geographic location candidates from an image of an environment are described. Open-source data indicative of the Earth's surface may be obtained and compared with images obtained from online sources. The images may be automatically analyzed using a plurality of modular convolution neural networks to determined probabilities of interest, environment, and if the image is locatable. Further, the resulting images may be analyzed for skyline and ridgeline depth orders and Region of Interest. A geolocation depicted in the image may be determined by comparing the results of the analysis with global geographic data.

First claim

Having thus described various embodiments of the invention, what is claimed as new and desired to be protected by Letters Patent includes the following:

1 . A method of determining a geographic location from an image, the method comprising:
obtaining one or more images, wherein a location where the one or more images was taken are unknown; determining a skyline in the one or more images; determining an elevation of the skyline; determining a depth of the skyline; determining one or more ridgelines in the one or more images and a ridgeline depth of each ridgeline of the one or more ridgelines; determining a mean depth ridgeline and a standard deviation of the ridgeline depth of the one or more ridgelines in the one or more images; determining at least one ridgeline by grouping the one or more ridgelines in each image of the one or more images based on the mean depth ridgeline and the standard deviation; and determining the location where the image was taken based at least in part on the elevation of the skyline, the depth of the skyline, and the at least one ridgeline.

2 . The method of claim 1 , wherein the one or more images is obtained from a plurality of images from online resources, and the skyline includes a plurality of features.

3 . The method of claim 1 , further comprising:
determining a first probability indicative of the one or more images depicting an outdoor environment; determining a second probability indicative of the one or more images depicting characteristics from which the location can be determined; and combining the first probability and the second probability to generate a total probability, wherein the total probability is indicative of a likelihood of determining the location that the one or more images was taken.

4 . The method of claim 1 , further comprising:
estimating one or more ridgeline elevations of the one or more ridgelines; and determining the location further based at least in part on the one or more ridgeline elevations.

5 . The method of claim 1 , further comprising:
obtaining content of interest, wherein the content of interest is indicative of military or illegal activity.

6 . The method of claim 1 , further comprising:
iteratively determining the skyline to produce a fine-grain elevation of the skyline and a fine-grain depth of the skyline; and determining the location where the image was taken based at least in part on the fine-grain elevation of the skyline and the fine-grain depth of the skyline.

7 . The method of claim 1 , further comprising:
determining man-made objects in the one or more images; and determining the location that the one or more images was taken further based at least in part on the man-made objects.

8 . The method of claim 1 , further comprising:
determining objects obscuring the skyline; masking the objects; estimating the skyline obscured by the objects to obtain an estimated skyline; and determining the location that the one or more images was taken further based at least in part on the estimated skyline.

9 . One or more non-transitory computer-readable media storing computer-executable instructions that, when executed by at least one processor, perform a method of determining a geographic location from an image, the method comprising:
obtaining a plurality of images, wherein locations where the plurality of images were taken are unknown; filtering one or more images from the plurality of images, wherein the one or more images depict outdoor environments and content of interest; determining a skyline in the one or more images; determining an elevation of the skyline in the one or more images; determining a depth of the skyline in the one or more images; determining a ridgeline in the one or more images; estimating a ridgeline depth and a ridgeline elevation of the ridgeline; determining a mean depth and a standard deviation of the ridgeline depth of each of ridgelines in each of the one or more images; determining at least one ridgeline by grouping the ridgelines in each image based on the mean depth and the standard deviation; and determining a location where the one or more images was taken based at least in part on the elevation of the skyline, the ridgeline elevation, and the at least one ridgeline.

10 . The media of claim 9 , wherein the method further comprises:
determining a probability of determining the location for the one or more images; and determining the one or more images from the probability.

11 . The media of claim 10 ,
wherein the probability is above a minimum threshold, wherein the method further comprises ordering the one or more images from a highest combined probability to a lowest combined probability to prioritize higher probability images.

12 . The media of claim 9 , wherein the one or more images are obtained from online resources.

13 . The media of claim 9 , wherein the method further comprises:
determining other objects in the one or more images; and determining a probability of determining the location of the one or more images based on the other objects and characteristics of the outdoor environments.

14 . The media of claim 13 , wherein the method further comprises:
filtering a set of images from the one or more images based on the probability of determining the one or more images; and determining the location for each image of the set of images based at least in part on the elevation of the skyline, the elevation of the ridgeline, the depth of the skyline, the depth of the ridgeline, and the other objects.

15 . The media of claim 14 , wherein the other objects are man-made objects.

16 . A method of determining a geographic location from an image, the method comprising:
obtaining a plurality of images, wherein locations where the plurality of images were taken are unknown; filtering a set of images from the plurality of images, wherein the set of images depict outdoor environments and content of interest; determining outdoor characteristics of the outdoor environments in the plurality of images; determining a probability of determining each location at which each image of the plurality of images was taken based at least in part on the outdoor characteristics; filtering a subset of images from the set of images, wherein each image of the subset of images is associated with a likelihood above a threshold of determining the locations where each image of the subset of images was taken; determining a skyline in each image of the subset of images; determining an elevation of the skyline in each image; determining a depth of the skyline in each image; determining a mean depth and a standard deviation of a ridgeline depth of ridgelines in the subset of images; determining at least one ridgeline by grouping the ridgelines in each image based on the mean depth and the standard deviation; and determining each location where each image of the subset of images was taken based at least in part on the elevation of the skyline, the depth of the skyline, and the at least one ridgeline.

17 . The method of claim 16 , further comprising ordering the subset of images from a highest combined probability to a lowest combined probability, wherein the lowest combined probability is above the threshold.

18 . The method of claim 16 , further comprising:
determining other objects in the plurality of images; and determining the probability of determining each location of each image further based on the other objects and the outdoor characteristics of

Assignees

• Applied Res Associates Inc

Inventors

• Warnaar Dirk B
• Besaw Lance E

Classifications

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06V20/13

  Satellite images · CPC title

• G06V20/30

  in albums, collections or shared content, e.g. social network photos or video · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06V10/82

  using neural networks · CPC title