Available Knowhow
Enables estimation of presence of noise as well as amount of noise
|
Categories |
Imaging / Computer Graphics, Digital Image Sensors |
|
Development Stage |
Working prototype |
|
Knowhow |
Available Knowhow |
|
Market |
Image-processing applications range from industrial uses and security systems to transportation and medical technology |
Highlights
-
All digital cameras, digital scanners and similar devices must estimate amount of noise in captured image. If the captured image is noisy the imaging device processes it to remove noise; if the captured image is clean the device must refrain from processing it.
Our Innovation
Method for estimating the amount of noise in an image captured by a sensor allows the device to estimate not just whether or not an image is noisy, but also provides an accurate estimate of the amount of noise present.
Key Features
-
Results of noise estimation were significantly more accurate than the state-of-the-art
-
Noise estimation on complex image with many edges and texture data (below). With low noise standard deviation other methods perform poorly; this method much better.
The Opportunity
-
Image-processing systems are used in quality control, industrial inspection, metrology for visually guided machines and to recognise components, symbolic characters and codes; robotics; driver assistance systems
1586128900.png)