Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part...

38

Transcript of Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part...

Page 1: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.
Page 2: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Region Detection

Defining regions of an image

Page 3: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.
Page 4: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Introduction All pixels belong to a region

Object Part of object Background

Find region Constituent pixels Boundary

Page 5: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Region Detection A set of pixels P An homogeneity predicate H(P) Partition P into regions {R}, such

that

iRi1

n

P

H iR True

H iR jR False

Page 6: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Point based methods – thresholding If

Regions are different brightness or colour

Then Can be differentiated using this

Page 7: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Global thresholds Compute threshold from whole image

Incorrect in some regions

Page 8: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Local thresholds Divide image into regions Compute threshold per region Merge thresholds across region

boundaries

Page 9: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Region Growing All pixels belong to a region Select a pixel Grow the surrounding region

Page 10: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Slow Algorithm If a pixel is

Not assigned to a region Adjacent to region Has colour properties not different to

region’s Then

Add to region Update region properties

Page 11: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Split and Merge Initialise image as a region While region is not homogeneous

Split into quadrants and examine homogeneity

Page 12: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Recursive SplittingSplit(P){ If (!H(P)) {

P subregions 1 … 4;Split (subregion 1);Split (subregion 2);Split (subregion 3);Split (subregion 4);

}}

Page 13: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Recursive Merging If adjacent regions are

Weakly split Weak edge

Similar Similar greyscale/colour properties

Merge them

Page 14: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Edge Following Detection

Finds candidate edge pixels Following

Links candidates to form boundaries

Page 15: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

4/8 ConnectivityProblem

Page 16: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Contour Tracking Scan image to find first edge point Track along edge points

Spurs? Endpoints?

Join edge segments

Page 17: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Edge Linking Aggregate collinear chain codes

Colinear?• Sequential least squares• tolerance band

Page 18: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Sequential Least Squares Accumulate best fitting line to segments and

error When error exceeds a threshold, finish

segment

Tolerance Band Accumulate best fitting line to segments If new point lies more than from line, finish

segment

Page 19: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

1. Take the first k edges from the list 2. Fit a line segment between the first and last 3. If the normalised maximum error is too large, shorten the sublist to the point of maximum error and return to step 2. 4. If the fit succeeds, compare this and the previous segments, if they are colinear, join them. 5. Advance the window of edges to get another k edges in the sublist and return to step 2.

Hop Along Algorithm

Page 20: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Examples An example would show an edge

detected image There would be a record of the

edge points constituting each edge segment

Page 21: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Scale Based Methods Structures observed depend on

scale of observation

Page 22: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Analysis Processing of an image should be

at a level of detail appropriate to structures being sought Image pyramid Wavelet transform

Page 23: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Image PyramidReducing resolution

Pixels in each layer computed by averaging groups of pixels in layer below. OrUse scale dependent operators – e.g. Marr Hildreth.

Page 24: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Wavelet Transform Transform image data Select coefficients Reverse transform

Page 25: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Watersheds of Gradient Magnitude Compare geographical watersheds

Divide landscape into catchment basins

Edges correspond to watersheds

Page 26: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Algorithm Locate local minima Flood image from these points When two floods meet

Identify a watershed pixel Build a dam Continue flooding

Page 27: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Examplewatersheds

local minima

Page 28: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.
Page 29: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

watershed point

Page 30: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

watershed point dam

Page 31: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.
Page 32: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Representing Regions Constituent pixels Boundary pixels

Page 33: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Region map As an array of region labels

Pixel value = region label

Page 34: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.
Page 35: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.
Page 36: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.
Page 37: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

Summary Region detection

Growing Edge following Watersheds

Page 38: Region Detection Defining regions of an image Introduction All pixels belong to a region Object Part of object Background Find region Constituent pixels.

I think there is a world market for maybe five computersThomas J Watson, chairman IBM, 1943