Calculate 3D Noise

Author: Dwight Urban Bartholomew
L3 Communications Infrared Products
dwight.bartholomew at
History: 2008/02/08: First version
Requires: ImageJ 1.39r
Installation: Download Calculate_3D_Noise.class to the plugins folder, or subfolder, restart ImageJ, and there will be a new "Calculate 3D Noise" command in the Plugins menu, or submenu.
This ImageJ plugin calculates the 3D-noise from a sequence
of images. 3D noise is a technique for describing the noise in
a video that tries to break down the analysis into various
components which helps to determine where the noise comes
from. For example, flicker noise (the t-component) is when
the average brightness of the image jumps around. Or rain
noise (the th-component) is like what was used in the Matrix
movies. Various ratios (say, vh/tvh) are general indicators
of the niceness of a video system.

The 3D-noise algorithm follows that developed by
the Night Vision Lab (or Night Vision Electronic 
Sensor Directorate) at Fort Belvoir, VA.

See: "Infrared Focal Plane Noise Parameter Definitions"
by J. G. Zeibel & R.T. Littleton at NVESD, October 2003

  1) Open an image sequence or video file and selects an ROI.
  2) Run this plug-in
  3) Verify the ROI.
  4) Input a Signal Transfer function (SiTF).
  This plugin was developed to analyze the 3D-noise
  of infrared images.  Thus, the SiTF has units to 
  convert image signal to degrees-Centigrade.  
Output in Results window includes:
  a) Eight 3D-noise components and the temporal noise.
     If an SiTF is entered, the units are milli-Kelvin.
  b) The SiTF (if used)
  c) The mean image signal within the ROI
  d) Size of the "Data Cube": height, width, # frames

|Plugins | Home |