Noise reduction with multi-frame super resolution
AssigneePixelworks, Inc.
InventorsGuodong Liu, Junhua Chen
Life of this patent
Estimating the term…
Abstract
A method of generating super resolution image data includes receiving original image data of a low resolution at an image processing device, performing motion compensation on the original image data using a current frame of image data and at least one previous frame of image data and at least one future frame of image data as reference frames, generating motion vectors, applying noise reduction to the current frame of image data and the reference frames to produce noise reduced, current frame image data, and generating a current frame of super resolution image data using the noise reduced, current frame image data.
Claims10
A method of estimating a noise level for a frame of image data, comprising:
- estimating an amount of noise at a pixel, wherein the amount of noise is calculated based on a difference between each interpolated pixel location in a current frame and a pixel in a noise reduced previous frame as previous frame differences, and a difference between each interpolated pixel location in the current frame and a pixel in a future frame as future frame differences;
- adjusting the previous and future frame differences based upon sample errors between pixels in the current frame to produce adjusted previous and adjusted future frame differences, wherein the sample error for each pixel in the current frame is a minimum distance between the interpolated pixel location and pixels surrounding the interpolated pixel location;
- calculating the noise level in the image data based upon a distribution of the noise from previous frame differences and a distribution of the noise from future frame differences to produce a previous noise level value and a future noise level value;
- and
- combining the previous and future noise level values to produce a combined noise level estimate for the current frame.
Dependencies are inferred from the claim text.