Structure from motion (SfM) processing for unmanned aerial vehicle (UAV)

Patent metadata

Field	Value
Publication number	US-10203692-B2
Application number	US-201515327917-A
Country	US
Kind code	B2
Filing date	Jul 10, 2015
Priority date	Jul 21, 2014
Publication date	Feb 12, 2019
Grant date	Feb 12, 2019

Abstract

A method of imaging an area using an unmanned aerial vehicle (UAV) collects a plurality of images from a sensor mounted to the UAV. The plurality of images are processed to detect regions that require additional imaging and an updated flight plan and sensor gimbal position plan is created to capture portions of the area identified as requiring additional imaging.

First claim

The invention claimed is:

1. A method of imaging an area using an unmanned aerial vehicle (UAV), the method comprising:
collecting a plurality of images from a sensor mounted to the UAV;
processing the plurality of images to detect regions that require additional imaging, wherein processing the plurality of images includes generating a point-cloud reconstruction of the area being imaged and comparing point density of areas to a threshold value wherein a point density of an area less than the threshold value indicates further imaging is required; and
creating a flight plan and sensor gimbal position plan to capture portions of the area identified as requiring additional imaging.

2. The method of claim 1 , further including:
receiving from the UAV positional and orientation information that describes the position and orientation of the UAV; and
associating the UAV positional and orientation information with each image collected by the sensor mounted to the UAV.

3. The method of claim 2 , further including:
receiving from the UAV gimbal orientation information that describes the orientation of the camera gimbal assembly; and
associating the UAV gimbal orientation information with each image collected by the sensor mounted to the UAV.

4. The method of claim 3 , wherein processing the plurality of images includes utilizing UAV positional and orientation information and gimbal orientation information associated with each image to detect images that overlap with one another, wherein overlapping images are grouped together.

5. The method of claim 4 , wherein processing the plurality of images further includes only processing images determined to overlap with other images.

6. The method of claim 5 , wherein processing the plurality of images further includes processing in parallel groups of images determined to overlap with one another.

Assignees

• Univ King Abdullah Sci & Tech

Inventors

• Smith Neil G
• Shalaby Mohamed
• Passone Luca

Classifications

• G01C11/025

  by scanning the object · CPC title

• B64C39/024

  of the remote controlled vehicle type, i.e. RPV · CPC title

• B64C2201/123

  Operations & Transport · mapped topic

• B64C2201/127

  Operations & Transport · mapped topic

• G01C21/165

  combined with non-inertial navigation instruments · CPC title