18038580. PERFORMING DENOISING ON AN IMAGE simplified abstract (KONINKLIJKE PHILIPS N.V.)

From WikiPatents
Jump to navigation Jump to search

PERFORMING DENOISING ON AN IMAGE

Organization Name

KONINKLIJKE PHILIPS N.V.

Inventor(s)

[[:Category:NIKOLAS DAVID Schnellb�cher of LÜBECK (DE)|NIKOLAS DAVID Schnellb�cher of LÜBECK (DE)]][[Category:NIKOLAS DAVID Schnellb�cher of LÜBECK (DE)]]

CHRISTIAN Wuelker of HAMBURG (DE)

FRANK Bergner of HAMBURG (DE)

KEVIN MARTIN Brown of CHARDON OH (US)

MICHAEL Grass of BUCHHOLZ IN DER NORDHEIDE (DE)

PERFORMING DENOISING ON AN IMAGE - A simplified explanation of the abstract

This abstract first appeared for US patent application 18038580 titled 'PERFORMING DENOISING ON AN IMAGE

Simplified Explanation

The patent application describes a mechanism for generating a partially denoised image by blending a weighted residual noise image with the original image using a convolutional neural network.

  • The residual noise image is obtained by processing the original image with a convolutional neural network.
  • The residual noise image is weighted before being combined with the original image to generate the partially denoised image.

Potential Applications

This technology could be applied in:

  • Image processing software for enhancing image quality.
  • Medical imaging for improving diagnostic accuracy.

Problems Solved

This technology addresses the following issues:

  • Noise reduction in images without losing important details.
  • Enhancing image quality for various applications.

Benefits

The benefits of this technology include:

  • Improved image quality.
  • Enhanced visual clarity.
  • Increased accuracy in image analysis.

Potential Commercial Applications

A potential commercial application for this technology could be:

  • Integration into photography software for professional photographers.

Possible Prior Art

One possible prior art for this technology could be:

  • Existing denoising algorithms used in image processing software.

Unanswered Questions

How does this technology compare to traditional denoising methods?

This technology uses a convolutional neural network to generate a partially denoised image by blending a weighted residual noise image with the original image. Traditional denoising methods may use filters or statistical techniques to reduce noise in images. The effectiveness and efficiency of this technology compared to traditional methods would be an important consideration for potential users.

What impact could this technology have on industries that rely heavily on image processing?

Industries such as healthcare, security, and entertainment rely heavily on image processing for various applications. The implementation of this technology could potentially revolutionize the way images are processed and analyzed in these industries. The impact on efficiency, accuracy, and overall performance would be crucial factors to consider for widespread adoption.


Original Abstract Submitted

A mechanism for generating a partially denoised image. A residual noise image, obtained by processing an image using a convolutional neural network, is weighted. The blending or combination of the weighted residual noise image and the (original) image generates the partially denoised image.

Retrieved from "http://wikipatents.org/index.php?title=18038580._PERFORMING_DENOISING_ON_AN_IMAGE_simplified_abstract_(KONINKLIJKE_PHILIPS_N.V.)&oldid=43840"