Segmentation H. Papasaika, E. Baltsavias Image Segmentation - - PowerPoint PPT Presentation
Segmentation H. Papasaika, E. Baltsavias Image Segmentation Partitioning of an image into a set of regions Regions representing meaningful areas of the image, such as crops, urban areas, forests. Various applications in Computer
Regions representing
meaningful areas of the image, such as crops, urban areas, forests.
Various applications in
Computer Vision and Geomatics.
“What is interesting and
what is not” depends on the application.
"I stand at the window and see a house, trees, sky. Theoretically I might say there were 327 brightnesses and nuances of colour. Do I have "327"? No. I have sky, house, and trees." -- Max Wertheimer Max Wertheimer
Gestalt theory: psychological theory of vision
Principle of Similarity:
Basic image Nearness Color similarity Shape similarity Common behavior
Haralick and Shapiro:
The regions should be uniform
uniform and homogeneous homogeneous with respect to some characteristic such as intensity value, color or texture.
Region interiors should be simple
simple and without many small holes holes.
Adjacent regions should have significantly different
different values values with respect to the characteristic on which they are uniform.
Boundaries of each segment should be simple
simple, not ragged ragged, and must be spatially accurate.
level knowledge
Segmentation is based on two basic properties
Discontinuity, i.e. to partition the image based on
abrupt changes in intensity (grey levels), e.g. edges
Similarity, i.e. to partition the image into similar
(according to predefined criteria) regions
Edge detection
Edge detection is a useful first step toward image segmentation, but not adequate by itself
Ideal case:
Segmentation techniques detecting intensity discontinuities
should yield pixels lying only on edges (or the boundary between regions).
Real life:
The detected set of pixels very rarely describes a complete edge
due to effects from: noise, breaks in the edge due to non-uniform illumination, similar object intensities etc.
Solution (partially):
Edge-detection techniques are followed by linking and other
boundary detection procedures which assemble edge pixels into meaningful boundaries.
Histogram Thresholding Feature Space Clustering Region-Based Approaches Edge Detection Approaches
Global Methods, e.g. K-means clustering Expectation Maximization (E/M) Local Methods, e.g. Watershed Segmentation Edge-based methods followed by region growing Simple thresholding
Light objects in dark
To extract the objects:
Select a threshold T
that separates the
background
i.e. any (x,y) for which
f(x,y) > T is an object point
A more general case of
this approach (multilevel thresholding)
So: pixel (x,y) belongs:
To one object if
T1< f(x,y) ≤ T2
To another if f(x,y) > T2 To another if
f(x,y) ≤ T1
A thresholded image:
(objects) (background) The chances of selecting a good threshold are increased if the histogram peaks are: Tall Narrow Separated by deep valleys
When T depends only on f(x,y)
When T depends on both f(x,y) and local regions
When T depends on x and y (in addition)
To partition the image histogram by using a
Then the image is scanned and labels are
This technique is successful in highly controlled
Incorrect in some regions
Divide image into regions Compute threshold per region Merge thresholds across region
Image Segmentation by Global Thresholding
Image Segmentation by Iterative Thresholding
Image Segmentation by Adaptive Thresholding
When a sensor makes available more than one
Each pixel is characterized by 3 values, and the
They involve algorithms that examine the
Identify connected groups of pixels in the
Agglomerative clustering
attach pixel to its closest cluster repeat
Divisive clustering
split cluster along “best” boundary repeat
Choose k pixels to act as cluster (class) centers Until the clustering is satisfactory (iterations)
Assign each pixel to the cluster that has the nearest
cluster center
Replace the cluster centers with the means of the
pixels in the clusters and continue iterations until cluster centers do not change significantly
K-means clustering using intensity alone and color alone
Image Clusters on intensity Clusters on color
Similarity measures can be chosen for the
Similarity according to intensity Similarity in color Similarity in texture Similarity in geometry
No good method of choosing similarity
– –Ri is a connected region, i = 1, 2, …, n –Ri ∩ Rj = 0 for all i and j, i≠j –P(Ri) = TRUE for i = 1, 2, …, n –P(Ri ⋃ Rj) = FALSE for i≠j
Segmentation is a process that partitions
n i i = =
1
P(Ri): logical predicate
Start with a set of “seed” points and from
Problems:
Seed selection Selection of suitable statistical criteria for
Sequence of processing Selection of homogeneity criterion Local vs. general criteria
Subdivide an image initially into a set of
Quadtree-based algorithm
If adjacent regions are
weakly split
weak edge, depending on defined criteria
similar
similar greyscale/colour properties
Merge them
Represent each pixel in a 1D
[luminance] or 3D [luminance,x,y] vector space.
Assume a fixed number k of
assumed to be a Gaussian cluster in this vector space.
Assume k initial centers and
covariances of clusters in this space.
Assign pixel i to cluster j with
weights corresponding to p(i|j).
Re-estimate the mean,
covariances of the clusters.
Loop steps 4, 5.
Flooding of the image gradient from pre-selected sources (waves emanate from set of markers)
The set of markers is grown until the exact contours of the objects are found at points where emanating waves meet (segmentation boundaries) Advantages : Robustness, Marker-based Flexibility Usage: Interactive and Automated Segmentation
Visualize an image in 3D: spatial coordinates and gray levels. In such a topographic interpretation, there are 3 types of points:
Points belonging to a regional minimum Points at which a drop of water would fall to a single minimum
(The catchment basin or watershed of that minimum)
Points at which a drop of water would be equally likely to fall
to more than one minimum. (The divide lines or watershed lines.)
Watershed lines
The objective is to find watershed lines The idea is simple:
Suppose that a hole is punched in each regional minimum and that
the entire topography is flooded from below by letting water rise through the holes at a uniform rate.
When rising water in distinct catchment basins is about to merge, a
dam is built to prevent merging. These dam boundaries correspond to the watershed lines = segment boundaries
A: Original image B: Negative of image A C: Distance transform of B D: Watershed transform of C
A B C D
Distance transform of a binary image is defined by the distance of every pixel to the nearest non-zero valued pixel
A: Original binary image C: Distance transform
a: Original image b: Gradient image of image a c: Watershed lines obtained from image b (oversegmentation) Each connected region contains one local minimum in the corresponding gradient image d: Watershed lines obtained from smoothed image b (due to smoothing no
a b c d
Watershed transform produces too many
One per local minimum Especially in noisy or highly detailed data
To alleviate oversegmentation
✔ Hierarchical approach – merge adjacent regions
Image T2 Preprocessed Image T2 Image T1 Preprocessed Image T1 Multiresolution Segmentation into Image objects Classification of changed objects Based on features from T1 and T2 Land Cover/ Land Use Change Maps
Image Processing Image Segmentation Impervious Surface Extraction Feature Extraction Attribution
Pan Sharpening Ortho Rectification eCognition Segmentation Fuzzy Logic Classification and Manual Cleanup Object Fusion to Create Features
customer Attribution of Polygons in ArcGIS
QuickbirdB, 2 feet GSD Before Segmentation After Segmentation After Segmentation With Object Boundaries
Segmentation to detect impervious surfaces
QuickBird Imagery 2 feet GSD Impervious Surface Feature Extraction
Segmentation of biomedical image (seeds and segments)
Coarse segmentation of image QuickBird satellite image 3.6 km
61 cm GSD
350 m
Sensitivity of segmentation to used parameters (undersegmenation left,
Segmentation example: Glacier
Segmentation example: Agricultural fields
Segmentation example: Agricultural fields
http://cmm.ensmp.fr/~beucher/wtshed.html http://www.umiacs.umd.edu/research/EXPAR/papers/
http://techtransfer.gsfc.nasa.gov/downloads/EUSC-
http://homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_C
http://www.owlnet.rice.edu/~elec539/Projects97/WDE
http://civs.stat.ucla.edu/Segmentation/Segment.htm