USA as Represented by Secretary of the Navy

Shibin Parameswaran

Bryan D Bagnall

Martin Thomas Jaszewski

Joshua Harguess

Corey A Lane

Gracie Bay Young

A method for processing images comprising: capturing a plurality of degraded images of a first real-world environment with a first sensor; processing each degraded image with a first, untrained convolutional neural network, via a Deep Image Prior approach, to obtain a plurality of clean images, wherein each clean image corresponds to a degraded image; pairing each clean image with its corresponding degraded image to create a plurality of degraded/clean image pairs; training, via a supervised learning approach, a machine learning model to learn a function for converting degraded images into restored images based on the plurality of degraded/clean image pairs; capturing a second plurality of degraded images of a second real-world environment; and using the trained machine learning model to convert the second plurality of degraded images into restored images based on the learned function.

Method for restoring images and video using self-supervised learning

A method for displaying off-screen target indicators in motion video comprising the steps of receiving motion video containing a series of individual video frames, selecting a target object within a selected video frame by choosing selected target object pixel space coordinates, and determining whether the selected target object pixel space coordinates are within the selected video frame. Upon determining that the selected target object pixel space coordinates are within the selected video frame, the method updates a dynamical system model with the target object geographical coordinates, longitudinal target object speed, and latitudinal target object speed. Upon determining that the selected target object pixel space coordinates are not within the selected video frame, the method calculates estimated target object geographical coordinates at time t using the dynamical system model. The method then calculates final values in the video field of view at which to draw a target indicator.

Method for displaying off-screen target indicators in motion video

Methods for detecting a horizon in an image with a plurality of pixels can include the step of blurring the image with a noise filter, then dividing the image into an M&#xd7;N matrix of sub-blocks S. For each sub-block S, horizon features can be coarse-extracted by defining an r-dimensional vector having P feature values for each sub-block S and clustering each r-dimensional vectors into two clusters using a k-means statistical analysis. The corresponding sub-blocks S corresponding to the two clusters can be masked with a binary mask. The methods can further include the step of fine-extracting the horizon features at a pixel level for each sub-block Sand sub-block Swhen the binary mask changes value from sub-block Sto said sub-block S, for i=1 to M and j=1 to N.

Growing community of inventors

San Diego, CA, United States of America

Shibin Parameswaran