User Guide

User Guide Neat Image
Copyright © 1999-2003 by ABSoft. All rights reserved.
16
20-Nov-03
5. Building device noise profiles
Each device noise profile describes the properties of noise produced by a device working in some mode.
Several device noise profiles corresponding to different device modes constitute a set that can be used to
process images produced in any of these device modes.
In this section, you will find out how to build a single device noise profile, how to document this profile
for future use, and how to prepare and document a structured set of profiles for your imaging device.
5.1. Building a device noise profile for specific device mode
Building a single device noise profile for specific device mode includes use of rough and fine-tuning
noise analyzers. Rough Analyzer is used to do initial analysis of image noise; it produces a rough
device noise profile. Fine-Tuning Analyzer improves the initial analysis to produce a fine-tuned device
noise profile.
Both rough and fine-tuning noise analyzers need uniform image areas to measure noise properties. If the
image has uniform areas that contain noise but no visible/important details, Neat Image can analyze the
noise properties using these areas. Neat Image itself cannot find the uniform areas (or areas that contain
no details important for you), so you have to specify areas that it should analyze.
Use the Device Noise Profile tab in the Filtration Job Editor:
5.1.1. Preparing a rough device noise profile
To build a rough device noise profile
1) Select working color space
⇒ Use the Working color space list in the Device Noise
Profile box.
The working color space is an internal parameter of the
filtration algorithm. An image is only temporarily
converted to the working color space for processing (the
input and output images are always in the RGB color
space).
We recommend using the YCrCb JPEG or YCrCb
Symmetric space to analyze and process color images.
The RGB color space is not the best for discerning image details that are masked by noise. Using
this color space most likely will not produce the best results.
Normally, you use the YCrCb JPEG (default) working color space for color photographic images or
the YCrCb Symmetric color space for grayscale (halftone) images. The RGB color space may also
be useful for specific applications. For example, if your intention is to filter only one specific color
channel (R, G or B) of the image then you use this space.
To use a specific working color space you need a noise profile built in that space. Neat Image will
re-build a noise profile if you change the working color space.
2) Find a uniform image area
⇒ Scroll, pan, zoom the input image to find a uniform area.
A uniform area (with minor variation in all channels)
may be overcast sky, clear sky (without clouds and
birds), or any other part of an image, where there are no visually perceptible details (except those
caused by noise).
The uniform area should preferable (but not necessary) be gray (neutral). The area should be at least
60x60 pixels large. That is the minimum size; the recommended size is 100x100 pixels or more.
See examples of uniform image areas
(
the subsection 10.1,
p
a
g
e 40
)