Table of Contents
Advertisement
From Analog to Digital
Analog to digital transformation. ppt 2-6.
Shows transformations from coninuous image, to
analog to digital.
For discussion on linear and non-linear cf ppt 2-10.
Processing the Digital Image
One of the most important advantages of digital imaging over screen/film is the abiltiy to enhance the diagnostic image by processing it, thereby improving its clinical
usefulness. The data are processed after the raw image is formed based on parameters set up in the system software. In the CR 800, the processing is performed by
specialized software in the INTERNAL PC.
A number of image processing techniques are possible, among them segmentation, tonescaling, and edge enhancement which are performed by the CR 800.
Additionally, users have some opportunity to further process the image by applying special features for improving qualities such as brightness (level) and contrast
(window).
1234563 – 12/12/94
Converting the Image Data
The next step is converting the analog signal into a digital signal by sampling the
blue light that comes off the plate and passing it through an analog-to-digital
converter to produce a binary (digital) value for the intensity of each sample.
Sampling is essentially taking a snapshot of the signal at a particular instant. The
sample has both a horizontal and a vertical dimension. The size of the sample is
predefined in the system software, both for the horizontal and vertical directions.
• The horizontal dimension represents a discrete instant in the movement of
the laser beam across the storage phosphor screen.
• The vertical dimension represents a discrete line on the screen perpendicular
to the scan.
If you were to locate both the horizontal and the vertical points of the sample on
an imaginary grid, as shown in the illustration, the result would represent a single
"pixel" in the final image.
Continuous anlog input values are converted to discrete output values, or
digitized signals.
In this process, small ranges of contiguous analog input values are replace by a
single digital output (code) value which represents one pixel of infomation on the
screen.
The output is a linear digital signal (the CR 800 produces a 16 bit digital signal
with a total signal range of 65,536 levels). This is a difficult range to work with
and impossilbe for the human eye to discriminate this range of values. Therefore,
the 16 bit linear image data is converted to12 bit log data (at least for the CR 800).
A 12 bit log gives a much more useful range of from 0 to 4095 values (4096 total).
The pixel (code) values of the CR 800 are mapped to this scale.
Computed Radiography (CR) Theory
2–15
Advertisement
Table of Contents
