Ch1-part2 ( Colour image processing )

Definitions :

Full colour The real colour present in nature like colour sensor needs in CCTV
camera
Pseudo colour The colour is assigned to specific monochrome intensity or
combination of intense
Radiance: The total amount of energy that flows from the light source (units:
watts)
Luminance: The amount of energy an observer can perceive from the light
source (lumens)
Brightness: The achromatic notion of image intensity.
Colour space or refers to a coordinate system where each colour stands for a
colour model: point.

Chromaticity: consists of hue and saturation

Hue: represents the dominant colour associated with the dominant
wavelength in range of light waves
Saturation: the degree of white light mixed with a hue
Colour space Refer to coordinate system where each colour stands for a point
Image Operation based on the colour differentiation, it is grouping
segmentation process that enables image pixel to be separated according to
their colour intensities
K-means is a technique that aims to partition observations into a number of
segmentation: clusters where each observation belongs to the cluster with the
nearest mean
Mean-shift clustering algorithm is to find the local maxima of the probability
segmentation: density given by the observations
Lossless These methods aim to retain lower compression ratios but
compression: preserve all the pixels in the original image
Lossy These approaches retain higher compression rates but sacrifice
compression: with a less resolution in the final compressed image

Colour image processing is divided into two main areas:

-full colour
-pseudo-colour processing
CCTV camera use
full colour sensor
There are three quantities that are used to describe the quality of a chromatic light
source: (list – define)
radiance-
- luminance and
Brightness-

To distinguish between two different colours, there are three essential

parameters: (list – define)
-brightness
hue -
-saturation

colour

Brightness Chromaticity

hue saturation

What is RGB?
-colour model system
-consist of three colours : Red, Green, Blue
-commonly use in image processing
-include large number of 24 bit in RGB cube

The often used colour models consist of

-RGB (red, green abd blue) model
-CMY (cyan, magentia and yellow) model
- CMYK (cyan, magenta, yellow and black) model
- HIS (hue, saturation and intensity) model

Why pseudo colour use in image processing?

Because Full-colour image processing is more complex than the pseudo-colour due
to the three colour vectors R , G ,B.
What is the different between smoothing and sharpening?

smoothing and sharpening are two basic manipulation tools on colour image processing,
smoothing, and sharpening are reverse operation

smoothing used to reduce the noise in the images

sharpening use to reproducing image by adding more details on the image

There are two types of segmentation:

- soft thresholding
That consist of - K-means segmentation\- mean shift segmentation
- hard thresholding

Soft Thresholding

1- K-means segmentation
-Technique that aims to partition observations into a number of clusters where
each observation belongs to the cluster with nearest mean

2- Mean shift segmentation

Segmentation that find the maxima of the probability density given by observation

The algorithm of the mean shift segmentation is followed:

-Start from a random region;
-Determine a centroid of the estimates;
-Continuously move the region towards the location of the new centroid;
-Repeat the iteration until convergence.
What is image compression ?
Reducing image size
There are two ways to compress colour image:
-lossless compression : compress the image (less compression) but preserve all the
pixels in the original image (high image quality – large space )
-lossy compression : high compression for the image ,but with less resolution (less
image quality)
True or false
1 \ one technique can be use of pseudo colour is image intensity slicing (true)

2 \ Full-colour image processing is more complex than pseudo-colour case due to

the three colour vectors (true)

3 \ smoothing and sharpening are two reverse process (true)

4 \ mean shift segmentation is a technique that aims to partition observations into

a number of clusters where each observation belongs to the cluster with the
nearest mean (false)

5 \ Each image pixel is formed by a number of bit,the number of these bits is namly
pixel depth (true)

5 \ image segmentation can easily leads to mistaken grouping outcomes (true)

Good luck

Partial dona: Dana Alhajri

Completed by Dr.Abeer