Bio-Rad Quantity One 1-D Analysis Software User Manual
Page 87
Quantity One User Guide
3-36
caused by dust or lint in the imaging optics or scratches on photographic film.
Salt is a type of outlier noise (see below).
•
Pepper. This type of noise appears as specks that are darker than the surrounding
background. The distribution histogram of this type of noise displays noise peaks
at the low end of the range (left end of the plot). Its causes are similar to those of
salt noise. Pepper is a type of outlier noise (see below).
Next, select one of the following option buttons to describe additional features of the
noise.
•
Gaussian. The distribution histogram of this type of noise has a Gaussian profile,
usually at the bottom of the data range. This type of noise is usually an electronic
artifact created by cameras and sensors, or by a combination of independent
unknown noise sources.
•
Uniform noise. This type of noise appears in the histogram as a uniform layer of
noise across the data range of the image.
•
Outlier noise. This category of noise includes salt and pepper noise (see above).
The distribution histogram of this type of noise displays noise peaks at the high
and low ends of the range.
After you have identified the type of noise, go to the next step.
Step 2: Select Filter Size
Image noise is filtered by means of a filtering window (or kernel), which is measured
in pixels. This filtering window slides across the image, processing the pixels within
it.
The available filter dimensions range from 3 x 3 to 9 x 9 pixels. To select an
appropriate size, magnify a background region of the image so that you can see the
individual pixels. The filter size you select should be larger than the average noise
feature but smaller than the data features.
Note:
A smaller filter will alter the image less than a larger filter. Large filters can result
in better suppression of noise, but can also blur desirable features in the image.