03-08-2012, 04:09 PM
Image Features and its Representation
introduction_image features.pptx (Size: 1.6 MB / Downloads: 26)
Representation
Once we extract the necessary features from a given image, we need to store them efficiently.
How we are going to store these features is called its Representation.
A feature class can have more than one representation associated with it.
Edge Representation
There are various techniques using which we can store the information about edges.
1.)Simple representation
2.)Chain codes
3.)Slope representation
4.)Signatures
Simple Representation
The simplest representation of a contour is using an ordered list of its edge points.
Advantages:
very useful for polygon shapes.
Give exact image when converted into images.
Disadvantages:
Not very effective representation for subsequent image analysis.
Not useful for representing curves.
Slope Representation
Slope representation
It is used for arbitrary directions.
(1) start at the first edge point and go clockwise around the contour.
(2) estimate the slope and arc length.
(3) plot the slope versus the arc-length.