+ All Categories
Home > Technology > Segmentation

Segmentation

Date post: 15-Jan-2015
Category:
Upload: guest49d49
View: 3,864 times
Download: 4 times
Share this document with a friend
Description:
 
Popular Tags:
12
Image Segmentation: Segmentation refers to the process of partitioning a digital image into multiple regions (sets of pixels). The goal of segmentation is to simplify or change the representation of an image into something that is more meaningful and easier to analyze. Image segmentation is typically used to locate objects and boundaries in images Each of the pixels in a region are similar with respect to some characteristic or computed property, such as color, intensity, or texture. Adjacent regions are significantly different with respect to the same characteristic Some applications of image segmentation in medical field includes: Locate tumors and other pathologies, Measure tissue volumes, Computer-guided surgery
Transcript
Page 1: Segmentation

Image Segmentation:• Segmentation refers to the process of partitioning a digital image into

multiple regions (sets of pixels).• The goal of segmentation is to simplify or change the representation of an

image into something that is more meaningful and easier to analyze.• Image segmentation is typically used to locate objects and boundaries in

images• Each of the pixels in a region are similar with respect to some characteristic

or computed property, such as color, intensity, or texture. • Adjacent regions are significantly different with respect to the same

characteristic• Some applications of image segmentation in medical field includes:

– Locate tumors and other pathologies, – Measure tissue volumes, – Computer-guided surgery

Page 2: Segmentation

What is Image Segmentation• Image segmentation reduces pixel data

to region-based information• segmentation of an image which

classifies voxels/pixels into objects or groups

• Form of segmenting the foreground

from background

• simpliest case: thresholding gray-scale pixel values

Page 3: Segmentation

Edge Detection:• Edges in images are areas with strong intensity contrasts –

a jump in intensity from one pixel to the next. • Edge detecting an image significantly reduces the amount

of data and filters out useless information, while preserving the important structural properties in an image.

• There are many ways to perform edge detection.– Gradient - The gradient method detects the edges by looking for

the maximum and minimum in the first derivative of the image. – Laplacian - The Laplacian method searches for zero crossings in

the second derivative of the image to find edges.

Page 4: Segmentation

Determining Intensity Values for Threshold

Thresholding separate foreground pixels from background pixels and can be performed before or after applying a morphological operation to an image. While a threshold operation produces a binary image and rely upon the definition of an intensity value.

This intensity value is compared to each pixel value within the image and an output pixel is generated based upon the conditions stated within the threshold.

Page 5: Segmentation

Intensity histograms provide a means of determining useful intensity values as well as determining whether or not an image is a good candidate for thresholding or stretching.

Intensity histogram based segmentation

Page 6: Segmentation

REGION GROWING• Group pixels or sub-regions into

larger regions when homogeneity criterion is satisfied

• Region grows around the seed point based on similar properties (grey level, texture, color)

PROS:• Better in noisy image where

edges are hard to identifyCONS:• Seed point must be specified• Different seed point will give

different results

PIXEL AGGREGATION:

Homogeneity criteria:• The difference between 2 pixel

values is less than or equal to 5• Horizontal, vertical, diagonal

10 10 10 10 10 10 10

10 10 10 69 70 10 10

59 10 60 64 59 56 60

10 59 10 60 70 10 62

10 60 59 65 67 10 65

10 10 10 10 10 10 10

10 10 10 10 10 10 10

Page 7: Segmentation

7

Region-Oriented Segmentation Region Splitting

Region growing starts from a set of seed points. An alternative is to start with the whole image as a single region and

subdivide the regions that do not satisfy a condition of homogeneity. Region Merging

Region merging is the opposite of region splitting. Start with small regions (e.g. 2x2 or 4x4 regions) and merge the

regions that have similar characteristics (such as gray level, variance). Typically, splitting and merging approaches are used iteratively.

Page 8: Segmentation

Split and Merge Approach:• This is a 2 step procedure:

– top-down: split image into homogeneous quadrant regions

– bottom-up: merge similar adjacent regions

• The algorithm includes:Top-down– successively subdivide image into

quadrant regions Ri

– stop when all regions are homogeneous: P(Ri ) = TRUE) obtain quadtree structure

Bottom-up– at each level, merge adjacent regions

Ri and Rj if P(Ri [ Rj ) = TRUE• Iterate until no further

splitting/merging is possible

Page 9: Segmentation

CONTOUR TRACING• It is a technique that is

applied to digital images in order to extract their boundary

• To trace the contour of a given pattern

Page 10: Segmentation

CONTOUR TRACING TECHNIQUE

• Palvidi’s algorithm

Page 11: Segmentation

ARITHMETIC OPERARTIONS

X


Recommended