Detecting three dimensional (3D) changes based on multi-viewpoint images

Patent metadata

Field	Value
Publication number	US-12579741-B2
Application number	US-202418422070-A
Country	US
Kind code	B2
Filing date	Jan 25, 2024
Priority date	Jan 25, 2024
Publication date	Mar 17, 2026
Grant date	Mar 17, 2026

Abstract

A method of detecting three dimensional (3D) changes in sites, comprising receiving a plurality of images of a site captured from a plurality of viewpoints by image sensor(s) mounted on dynamic vehicle(s) scanning the site, registering at least some of the images to each other, creating one or more merged representations of at least part of the site based on the at least some registered images, comparing between the merged representation(s) and one or more corresponding previous merged representations created based on a plurality of previous images captured during one or more previous scans of the site, detecting, based on the comparison, one or more 3D changes indicative of a depth change exceeding a certain threshold in one or more 3D features at the site, and generating a change indication indicative of the one or more detected 3D changes.

First claim

What is claimed is:

1 . A method of detecting three dimensional (3D) changes in sites, comprising:
using at least one processor for: receiving a plurality of images of a site captured from a plurality of viewpoints by at least one image sensor mounted on at least one dynamic vehicle scanning the site; registering at least some of the plurality of images to each other; creating at least one merged representation of at least part of the site based on the at least some registered images depicting the at least part of the site from at least some of the plurality of viewpoints; comparing between the at least one merged representation and at least one corresponding previous merged representation created based on a plurality of previous images captured during at least one previous scan of the site; detecting, based on the comparison, at least one 3D change indicative of a depth change exceeding a certain threshold in at least one 3D feature at the site; and generating a change indication indicative of the at least one detected 3D change; wherein the at least one corresponding previous merged representation is selected according to geolocation metadata associated with each of a plurality of previous merged representations of the site, the geolocation metadata is indicative of the geolocation of the at least part of the site depicted in the plurality of previous images used to create the respective previous merged representation.

2 . The method of claim 1 , wherein the at least some images are registered each other according to at least one common visual feature detected in the at least some images.

3 . The method of claim 1 , wherein the at least one merged representation is created based on pixel subtraction of corresponding pixels in the at least some registered images.

4 . The method of claim 1 , wherein the at least one merged representation is created based on at least partial 3D reconstruction of the at least part of the site using the at least some registered images.

5 . The method of claim 1 , wherein the at least one merged representation comprises a point cloud model of the at least part of the site.

6 . The method of claim 1 , wherein the at least one merged representation comprises a textured 3D model of the at least part of the site.

7 . The method of claim 1 , wherein the at least one merged representation comprises an ortophoto of the at least part of the site.

8 . The method of claim 1 , wherein the at least one merged representation comprises at least one panoramic image of the at least part of the site created by stitching at least some of the plurality of images.

9 . The method of claim 1 , wherein the at least one 3D change is detected using at least one trained machine learning model.

10 . The method of claim 1 , wherein the at least one 3D change is detected using at least one image processing algorithm.

11 . A method of detecting three dimensional (3D) changes in sites, comprising:
using at least one processor for:
receiving a plurality of images of a site captured from a plurality of viewpoints by at least one image sensor mounted on at least one dynamic vehicle scanning the site;
registering at least some of the plurality of images to each other;
creating at least one merged representation of at least part of the site based on the at least some registered images depicting the at least part of the site from at least some of the plurality of viewpoints;
comparing between the at least one merged representation and at least one corresponding previous merged representation created based on a plurality of previous images captured during at least one previous scan of the site;
detecting, based on the comparison, at least one 3D change indicative of a depth change exceeding a certain threshold in at least one 3D feature at the site; and
generating a change indication indicative of the at least one detected 3D change;
wherein the at least one merged representation is registered to the corresponding previous merged representation according to at least one common visual feature detected in the at least one merged representation and in the corresponding previous merged representation.

12 . A method of detecting three dimensional (3D) changes in sites, comprising:
using at least one processor for: receiving a plurality of images of a site captured from a plurality of viewpoints by at least one image sensor mounted on at least one dynamic vehicle scanning the site; registering at least some of the plurality of images to each other; creating at least one merged representation of at least part of the site based on the at least some registered images depicting the at least part of the site from at least some of the plurality of viewpoints; comparing between the at least one merged representation and at least one corresponding previous merged representation created based on a plurality of previous images captured during at least one previous scan of the site; detecting, based on the comparison, at least one 3D change indicative of a depth change exceeding a certain threshold in at least one 3D feature at the site; and generating a change indication indicative of the at least one detected 3D change;
applying at least one reconstruction error detection algorithm to determine whether the at least one detected 3D change is induced by at least one reconstruction error in the at least one merged representation and/or in the at least one corresponding previous merged representation.

13 . A method of detecting three dimensional (3D) changes in sites, comprising:
using at least one processor for: receiving a plurality of images of a site captured from a plurality of viewpoints by at least one image sensor mounted on at least one dynamic vehicle scanning the site; registering at least some of the plurality of images to each other; creating at least one merged representation of at least part of the site based on the at least some registered images depicting the at least part of the site from at least some of the plurality of viewpoints; comparing between the at least one merged representation and at least one corresponding previous merged representation created based on a plurality of previous images captured during at least one previous scan of the site; detecting, based on the comparison, at least one 3D change indicative of a depth change exceeding a certain threshold in at least one 3D feature at the site; and generating a change indication indicative of the at least one detected 3D change;
first detecting at least one 2D change indicative of at least one potential 3D change in the at least part of the site by:
comparing between at least one of the plurality of images depicting the at least part of the site and at least one corresponding previous image of the plurality of previous images, detecting at least one 2D change based on the comparison, comparing between at least section of the at least one merged representation and the at least one corresponding previous merged representation in which the at least one 2D change is detected, and determining whether the at least one 2D change is a 3D change indicative of a depth change exceeding a certain threshold in at least one 3D feature at the site.

14 . The method of claim 13 , wherein a path followed of the at least one dynamic vehicle to scan the site is similar to a path followed during the at least one previous scan.

15 . The method of claim 13 , wherein the at least one dynamic vehicle is an aerial vehicle which is a member of a group consisting of: a drone, an unmanned aerial vehicle (UAV), and a

Assignees

• Nec Corp America

Inventors

• Lev Tsvi
• Hoch Yaacov
• Machluf Nisim
• Katz Or
• Kirshner Naomi Sara

Classifications

• G06V10/16

  using multiple overlapping images; Image stitching · CPC title

• G06V10/761

  Proximity, similarity or dissimilarity measures · CPC title

• G06V20/17 Primary

  taken from planes or by drones · CPC title

• G06T7/33

  using feature-based methods · CPC title

• G06T7/00

  Image analysis · CPC title