Back to results
US 2020/0092453 A1Published application
Photographic Underexposure Correction Using A Neural Network
AssigneeSpectrum Optix Inc.
InventorsKevin Gordon, Darcy Daugela, Martin Humphreys
Life of this patent
Estimating the term…
Abstract
A method for image capture includes determining an exposure range and setting at least one camera parameter to capture an underexposed image outside the exposure range. The underexposed image is processed using a neural network to recover image details. Image defects due to camera or object motion blur can be reduced.
Claims20
A method for image capture, comprising the steps of:
- determining an exposure range;
- setting at least one camera parameter to capture an underexposed image outside the exposure range;
- processing the underexposed image using a neural network to recover image details.
A method for image improvement, comprising the steps of:
- determining analog settings for a sensor system having a sensor;
- capturing data with the sensor;
- and
- using a neural network trained on previous data provided by the sensor to reduce noise and recover image details from the captured image data.
1 more element
A system for improving total system gain in combined analog and digital systems, comprising
- a sensor system having a sensor configured to capture image data, with the image data processed in a sequential analog and digital procedure;
- a neural network trained on previous data provided by the sensor or sensor type to reduce noise and recover image details from the captured image data after analog amplification.
Dependencies are inferred from the claim text.