Home Patent Forecast® Sectors Log In   Sign Up   Support   Contact  
Why Patent Forecast® What's Included Patent Forecast® Sectors Free Trial Pricing News Feed Subscribe Menu
Enjoy your FREE PREVIEW which shows only 2018 data and 25 documents. For full access, subscribe at any time.         Subscribe

Unmanned Aerial Vehicles

Search All Patents in Unmanned Aerial Vehicles


Application 20180181790


Published 2018-06-28

Object Detection System

An airborne mine countermeasure system includes a processor coupled to a memory having stored therein software instructions that, when executed by the processor, cause the processor to perform a series of image processing operations. The operations include obtaining input image data from an external image sensor, and extracting a sequence of 2-D slices from the input image data. The operations also include performing a 3-D connected region analysis on the sequence of 2-D slices, and extracting 3-D invariant features in the image data. The operations further include performing coarse filtering, performing fine recognition and outputting an image processing result having an indication of the presence of any mines within the input image data.


Classification


Slightly Less than Average Length Specification


View the Patent Matrix® Diagram to Explore the Claim Relationships

4 Independent Claims

  • Independent Claim 1-20. (canceled)

  • Independent Claim 21. An object detection system, comprising: an image sensor configured to obtain image data in real-timea processor for executing modulesand a memory containing the modules, the modules including a live data processing system configured to process the image data in real-timean invariant feature extraction module configured to extract 3-D invariant features of at least one object in the image dataand a classifier configured to compare the 3-D invariant features of the at least one object in the image data to a data store of trained 3-D invariant features of known objects, wherein the classifier is further configured to output an indication of how close the 3-D invariant features of the at least one object in the image data match the trained 3-D invariant features.

  • Independent Claim 30. A computerized method for detecting objects, comprising: obtaining, at a processor, image data in real-time from an image sensorprocessing, using the processor, the image data in real-time by extracting 3-D invariant features of at least one object in the image datacomparing, using the processor, the 3-D invariant features of the at least one object in the image data to a data store of trained 3-D invariant features of known objectsand outputting an indication of how close the 3-D invariant features of the at least one object in the image data match the trained 3-D invariant features.

  • Independent Claim 37. A non-transitory computer readable medium having stored thereon instructions that, when executed by a processor, cause the processor to perform a series of operations to detect objects, the detecting including: obtaining, at a processor, image data in real-time from an image sensorprocessing, using the processor, the image data in real-time by extracting 3-D invariant features of at least one object in the image datacomparing, using the processor, the 3-D invariant features of the at least one object in the image data to a data store of trained 3-D invariant features of known objectsand outputting an indication of how close the 3-D invariant features of the at least one object in the image data match the trained 3-D invariant features.