Granted Patent B2

US 12,430,909 · App. 17/636,838 · Granted Sep 30, 2025

Inventors: Adam Murphy (Manchester, NH); Mark Johnson (Vannes, FR)

Assignee: Raymarine UK Limited

G06V20/17G01C21/20G05D1/0094G05D1/104G05D1/106G06V20/05B64U10/14B64U70/93B64U2101/30

Abstract

Claims (57)

1. An apparatus comprising:

an unmanned aerial system (UAS) comprising a flight platform configured to execute a search path to search for an underwater object as the UAS flies over a sea and adjusts a minimum flight altitude based on sea calmness or choppiness;

an image capture device comprising image capture components configured to generate a stream of images corresponding to a field of view of the UAS;

a logic device associated with the UAS and configured to:

analyze the stream of images using a marine video analysis (MVA) system to detect a region of interest comprising an underwater object based on identifying an object of interest on and/or above a water surface and an association of the identified object with a presence of the underwater object, wherein the MVA system is configured to identify the object of interest on and/or above the water surface based at least in part on a detected size and/or shape;

after a region of interest is detected, process the stream of images associated with the region of interest through a neural network trained to identify the underwater object in the detected region of interest identified by the MVA; and

notify a mobile structure of the identified underwater object;

wherein the sea calmness or choppiness are obtained from analysis of visual and/or thermal data captured by the image capture device.

2. The apparatus of claim 1 , wherein the logic device is further configured to:

determine a mobile structure route based, at least in part, on the notification of the identified object, including object location.

3. The apparatus of claim 2 , wherein the search path is an automated search path based, at least in part, on chlorophyl levels and/or water color and/or water clarity.

4. The apparatus of claim 1 , wherein the logic device is further configured to determine and communicate positioning instructions to the UAS, and wherein the positioning instructions are determined based, at least in part, on a movement vector of the mobile structure.

5. The apparatus of claim 1 , wherein the mobile structure is configured to receive UAS data from a plurality of UASs associated with the mobile structure, filter the UAS data to identify objects associated with one or more target fish species, and display at least a portion of the filtered UAS data as an overlay on sonar, radar, vessel location, and/or fishing location maps.

6. The apparatus of claim 1 , wherein the UAS is configured to optimize a flight altitude and/or camera angle to maximize the field of view while still ensuring a range that will provide a minimum number of pixels comprising target objects for MVA detection and neural network object identification.

7. The apparatus of claim 5 , wherein the logic device is further configured to:

receive one or more search patterns, including environmental data including sun position, fog conditions, and/or sea conditions;

generate and transmit maneuvering and/or monitoring instructions for one or more of the UASs based on the received one or more search patterns, including orienting the UAS to mitigate solar glare, increase visibility, and/or avoid waves and/or swells; and

monitor the UAS data received from the one or more of the UASs as they traverse the one or more search patterns for a detected object.

8. The apparatus of claim 1 , wherein the neural network comprises a convolutional neural network configured to classify the object and generate an associated confidence value;

wherein the classified object includes one or more fish, floating debris, seaweed patches, birds, and/or vessels; and

wherein the confidence value indicates a probability of fish being present at a location associated with the classified object;

wherein the logic device is further configured to:

determine that the associated confidence value is below a threshold; and

generate maneuvering instructions for the UAS to depart from the search path to capture higher quality images of the region of interest in response to determining that the associated confidence value is below the threshold.

9. The apparatus of claim 1 , wherein the logic device is further configured to automatically notify the mobile structure after the underwater object is detected, the notification including the captured image, object identification, object location and/or other environmental data comprising date and time, global positioning satellite location, sea surface temperature, fish location and depth, fish type and size, bottom density and composition, and/or bird flock size.

10. A method comprising:

capturing a stream of images from an image capture device coupled to an unmanned aerial system (UAS) associated with a mobile structure as the UAS flies over a sea and adjusts a minimum flight altitude based on sea calmness or choppiness;

detecting, by a marine video analytics system (MVA), a region of interest comprising an underwater object in the captured stream of images based at least in part on identifying an object of interest on and/or above a water surface and an association of the identified object with a presence of the underwater object, wherein detecting, by the MVA, further comprises analyzing the stream of images for the object on and/or above the water surface based at least on a size and shape of the object;

identifying, by a neural network trained to identify the underwater object from the output of the MVA, the underwater object in the region of interest, including an object classification and an associated confidence value;

determining that the associated confidence value is below a threshold;

generating maneuvering instructions for the UAS to capture higher quality images of the underwater object in response to determining that the associated confidence value is below the threshold; and

notify a mobile structure of the identified object if the associated confidence value exceeds the threshold;

wherein the sea calmness or choppiness are obtained from analysis of visual and/or thermal data captured by the image capture device.

11. The method of claim 10 , further comprising:

determining a mobile structure route based, at least in part, on a location of the underwater object.

12. The method of claim 10 , further comprising:

receiving UAS data from the UAS;

filtering the UAS data to identify objects associated with one or more target fish species; and

displaying at least a portion of the filtered UAS data received from the one or more UASs on a display of a user interface as an overlay on a sonar map, a radar map, a vessel location map and/or a fishing location map.

13. The method of claim 10 , further comprising:

receiving automated maneuvering and/or monitoring instructions for the UAS, including an automated search path based, at least in part, on chlorophyl levels; and

transmitting the received maneuvering and/or monitoring instructions to the UAS.

14. The method of claim 10 , further comprising:

receiving one or more search patterns, including environmental data including sun position, fog conditions, and/or sea conditions;

generating and transmitting maneuvering and/or monitoring instructions for the UAS based on the received one or more search patterns, including orienting the UAS to mitigate solar glare, increase visibility, and/or avoid waves and/or swells; and

monitoring data received from the UAS as the UAS traverses the one or more search patterns.

15. The method of claim 10 , further comprising:

launching the UAS from the mobile structure; and

optimizing a flight altitude and/or camera angle to maximize the field of view while still ensuring the minimum flight altitude.

16. The method of claim 14 , further comprising:

capturing the stream of images during execution of the one or more search patterns.

17. The method of claim 11 , wherein the neural network comprises

a convolutional neural network trained to identify the detected object and generate an associated confidence value;

wherein the identified object includes one or more fish, floating debris, seaweed patches, birds, and/or vessels; and

wherein the confidence value indicates a probability of fish being present at an underwater location associated with the classified object.

18. The method of claim 10 , further comprising:

communicating data from the UAS to the mobile structure including the captured image, object identification, object location and/or other environmental data comprising date and time, global positioning satellite location, sea surface temperature, fish location and depth, fish type and size, bottom density and composition, and/or bird flock size.

Assignments (2)

Continuity (2)

References Cited (26)