Mask Application

[Laboratory]

The mask or "validity mask" segment within an image file informs which pixels in the image are valid. There are several applications that use this facility or concept.

The data model with the "validity mask" segment is shown below.

Data model for mask operations

We will illustrate the use of it with an ultrasound image. Shown below is the ultasound image and we wish to compute the statistics of it. Note that there is a "black" area that contains no valid information.

Ultrasound image

Computing statistics on the image will include the entire "black" area.

  Object Dimension:  w=250  h=200  d=1  t=1  e=1
  Mean:     63.2069
  Variance: 2132.67
  Std Dev:  46.1808
  RMS:      78.2799
  Total Contributing Points:    50000

If we incorporate a validity mask then the statistics computed on the image will not include the "black" area, i.e., only the unmasked data will be used for the computation. Shown below are the validity mask and the image with the validity mask segment.

a)Validity mask; b)Image with validity mask
a) b)

The new statistics are shown below.

  Object Dimension:  w=250  h=200  d=1  t=1  e=1
  Mean:     77.2634
  Variance: 2384.87
  Std Dev:  48.8351
  RMS:      91.4026
  Total Contributing Points:    33816

Below, the histogram of the original image and the histogram of the data of interest.

a)histogram of the original image; b)histogram of the image with validity mask
a) b)





Main DIP Menu
DIP Feedback Form
Copyright © 1997-1995 KRI, ISTEC, Ramiro Jordán, Roberto Lotufo. All Rights Reserved