Masking Pixels

During data analysis it is at times important to exclude part of the data (=pixels) from further analysis. Epina ImageLab supports the temporary exclusion of pixels by means of up to 32 pixel masks, containing masking information for different purposes. Thus one mask may contain information leading to the exclusion of actual "bad pixels" (e.g. if the FPA detector exhibits dead pixels), another mask may mark all pixels which belong to the background, and so on.

Typical reasons for masking pixels include the removal of background pixels which just contribute noise, the exclusion of bad pixels (e.g. caused by spikes or by a broken detector), or the exclusion of a majority of pixels which originate from spots in the image which are currently of no interest.

Epina ImageLab offers several tools to identify a certain class of pixels, to copy them to a mask, and to manipulate the masks:

  • Mask Editor - this tool has mainly been designed to create pixels masks and edit them at the pixel level
  • The Mask Creator is a light-weight tool to quickly create masks from all kinds of processed image data. The mask creator can be opened by selecting the command "Create a Mask from Current Image Data" from the context menu of a particular image.
  • The Signal Distribution Analyser uses the distribution of intensities at particular wavelengths to derive pixel masks.
  • Suspicious Pixel Calculator: a tool to identify pixels which are "suspicious" in various ways.
  • The Epina ImageLab Script Language - Pixel Masks provides procedures for the manipulation of pixel masks.

The definition of pixel masks usually works along the following lines:

  1. Find and mark pixels
  2. Create, if necessary, one or more pixel mask(s)
  3. Copy/Add/Subtract marked pixels to/from an existing pixel mask
  4. Apply a particular pixel mask in connection with image analysis