11
Signals and Biomedical Signal Processing
50
100
150
200
250
300
350
400
450
500
50
100
150
200
250
300
350
400
450
500
(c)
FIGURE 1.4 (continued) (c) range 0–1.
Color images are also used in medical imaging. While there are many standards
for color images, here we discuss only “red green blue” or “RGB” standard. RGB is
formed based on the philosophy that each color is a combination of the three primary
colors: red, green, and blue. This means that if we combine the right intensity of these
three colors, we can create the sense of any desired color for the human eyes. As an
example, for a purple object we would have a high density of red and blue but a low
intensity of green. As a result, in RGB representation of a color image, the screen (such
as a monitor) provides three dots for every pixel (point): one red dot, one green dot,
and one blue dot. The intensity of each of these dots is identified by the share of the
corresponding primary color in forming the color of the pixel. This means that in color
images for every coordinate (x, y), three numbers are provided. This in turn means that
the image itself is represented by three 2-D signals, g R (x, y), g G (x, y), and g B (x, y), each
representing the intensity of one primary color. As a result, every one of the 2-D signals (for one color) can be treated as one separate image and processed by the same
image processing methods designed for gray-level images.
1.6.3 IMAGE HISTOGRAM
An important statistical characteristic of an image is the histogram. Here, we
define this concept and illustrate it using a simple example. Assume that the gray
level of all pixels in an image belong to the interval [0, G − 1], where G is an integer. Consequently, if “r” represents the gray level of a pixel of the image, then
0 ≤ r ≤ G − 1, where r is an integer. Now, for all values of r, calculate the normalized frequencies, p(r). In order to do so, for a given gray-level value r, we count the
Signals and Biomedical Signal Processing
50
100
150
200
250
300
350
400
450
500
50
100
150
200
250
300
350
400
450
500
(c)
FIGURE 1.4 (continued) (c) range 0–1.
Color images are also used in medical imaging. While there are many standards
for color images, here we discuss only “red green blue” or “RGB” standard. RGB is
formed based on the philosophy that each color is a combination of the three primary
colors: red, green, and blue. This means that if we combine the right intensity of these
three colors, we can create the sense of any desired color for the human eyes. As an
example, for a purple object we would have a high density of red and blue but a low
intensity of green. As a result, in RGB representation of a color image, the screen (such
as a monitor) provides three dots for every pixel (point): one red dot, one green dot,
and one blue dot. The intensity of each of these dots is identified by the share of the
corresponding primary color in forming the color of the pixel. This means that in color
images for every coordinate (x, y), three numbers are provided. This in turn means that
the image itself is represented by three 2-D signals, g R (x, y), g G (x, y), and g B (x, y), each
representing the intensity of one primary color. As a result, every one of the 2-D signals (for one color) can be treated as one separate image and processed by the same
image processing methods designed for gray-level images.
1.6.3 IMAGE HISTOGRAM
An important statistical characteristic of an image is the histogram. Here, we
define this concept and illustrate it using a simple example. Assume that the gray
level of all pixels in an image belong to the interval [0, G − 1], where G is an integer. Consequently, if “r” represents the gray level of a pixel of the image, then
0 ≤ r ≤ G − 1, where r is an integer. Now, for all values of r, calculate the normalized frequencies, p(r). In order to do so, for a given gray-level value r, we count the