Back to results
US 12,417,710 B2Granted
Unmanned aerial vehicle search and rescue systems and methods
AssigneeFLIR Unmanned Aerial Systems ULC
InventorsAhmed Elhossini, Fei Yu
Life of this patent
Estimating the term…
Abstract
Systems and methods for unmanned aerial vehicle (UAV) search and rescue operations are provided. A UAV may include a camera configured to capture one or more images of a search area. The UAV may include at least one logic device configured to calculate a flight pattern for searching the search area, receive a target image, and analyze the one or more images of the search area to identify a target matching the target image.
Claims20
An unmanned aerial vehicle (UAV) comprising:
- a body;
- a camera configured to capture one or more images of a search area;
- a gimbal configured to secure the camera to the body and point the camera independent of a position and/or orientation of the body;
- a plurality of navigation cameras configured to provide one or more images for a position estimation of the UAV, the plurality of navigation cameras comprising a front camera pointing in a direction of travel of the UAV and multiple side cameras pointing down and to the sides of the body;
- at least one logic device configured to:
- receive a selection of the search area to be searched;
- calculate a flight pattern for searching the search area based on at least the selection, wherein the flight pattern comprises a plurality of waypoints that cover the search area;
- receive a target image;
- and
- analyze the one or more images of the search area to identify a target matching the target image, wherein analyzing the one or more images comprises:
- calculating a template descriptor based on the target image;
- creating one or more patch descriptors associated with one or more patches of each image of the one or more images;
- comparing, using a template matching algorithm, the template descriptor against each of the patch descriptors to calculate a correlation between the template descriptor and each of the one or more patch descriptors;
- and
- identifying the target based on the correlation between the template descriptor and each of the one or more patch descriptors.
12 more elements
A system comprising:
- a user interface configured to receive user input selecting a search area;
- an unmanned aerial vehicle (UAV) in communication with the user interface, the UAV comprising a camera configured to capture one or more images of the search area;
- a flight module configured to calculate a flight pattern of the UAV for searching the search area based on at least the selection, wherein the flight pattern comprises a plurality of waypoints that cover the search area;
- and
- an imagery module configured to:
- receive a target image;
- and
- analyze the one or more images of the search area to identify a target matching the target image, wherein analyzing the one or more images comprises:
- calculating a template descriptor based on the target image;
- creating one or more patch descriptors associated with one or more patches of each image of the one or more images;
- comparing, using a template matching algorithm, the template descriptor against each of the patch descriptors to calculate a correlation between the template descriptor and each of the one or more patch descriptors;
- and
- identifying the target based on the correlation between the template descriptor and each of the one or more patch descriptors.
10 more elements
A method comprising:
- receiving a target image;
- receiving a selection of a search area to be searched;
- calculating a flight pattern of an unmanned aerial vehicle (UAV) for searching the search area based on at least the selection, wherein the flight pattern comprises a plurality of waypoints that cover the search area;
- capturing, using a camera of the UAV, one or more images of the search area;
- estimating, using a plurality of navigation cameras, a position of the UAV, wherein the plurality of cameras comprises a front camera pointing in a direction of travel of the UAV and multiple side cameras pointing down and to the sides of the body;
- and
- analyzing the one or more images to identify a target matching the target image, wherein analyzing the one or more images comprises:
- calculating a template descriptor based on the target image;
- creating one or more patch descriptors associated with one or more patches of each image of the one or more images;
- comparing, using a template matching algorithm, the template descriptor against each of the patch descriptors to calculate a correlation between the template descriptor and each of the one or more patch descriptors;
- and
- identifying the target based on the correlation between the template descriptor and each of the one or more patch descriptors.
9 more elements
Dependencies are inferred from the claim text.