Back to results
US 2020/0050893 A1Published application
Training a deep learning system for maritime applications
AssigneeBuffalo Automation Group Inc.
InventorsThiru Vikram SURESH, Mohit Arvind KHAKHARIA
Life of this patent
Estimating the term…
Abstract
An object detection network can be trained with training images to identify and classify objects in images from a sensor system disposed on a maritime vessel. The objects in the images can be identified, classified, and heat maps can be generated. Instructions can be sent regarding operation of the maritime vessel. For some training images, water conditions, sky conditions, and/or light conditions in the image can be changed to generate a second image.
Claims20
A method comprising:
- training, using a processor, an object detection network with training images to identify and classify objects in images from a sensor system disposed on a maritime vessel;
- identifying objects in the images using the processor in an offline mode;
- classifying the objects in the images using the processor in the offline mode
- generating heat maps in the offline mode;
- and
- sending instructions regarding operation of the maritime vessel, using the processor, based on the objects that are identified, wherein the instructions include a speed or a heading.
3 more elements
A method comprising:
- receiving an image of a maritime object at a processor;
- and
- changing at least one of water conditions, sky conditions, and light conditions in the image of the maritime object to generate a second image using the processor.
Dependencies are inferred from the claim text.