Unsupervised image segmentation is an important component in many
image understanding algorithms and practical vision systems. However,
evaluation of segmentation algorithms thus far has been largely subjective,
leaving a system designer to judge the effectiveness of a technique
based only on intuition and results in the form of a few example
segmented images. This is largely due to image segmentation being
an ill-defined problem-there is no unique ground-truth segmentation
of an image against which the output of an algorithm may be compared.
This paper demonstrates how a recently proposed measure of similarity,
the Normalized Probabilistic Rand (NPR) index, can be used to perform
a quantitative comparison between image segmentation algorithms using
a hand-labeled set of ground-truth segmentations. We show that the
measure allows principled comparisons between segmentations created
by different algorithms, as well as segmentations on different images.
We outline a procedure for algorithm evaluation through an example
evaluation of some familiar algorithms-the mean-shift-based algorithm,
an efficient graph-based segmentation algorithm, a hybrid algorithm
that combines the strengths of both methods, and expectation maximization.
Results are presented on the 300 images in the publicly available
Berkeley Segmentation Data Set.
%0 Journal Article
%1 Unnikrishnan2007
%A Unnikrishnan, Ranjith
%A Pantofaru, Caroline
%A Hebert, Martial
%D 2007
%J IEEE TPAMI
%K Algorithms,Artificial Automated,Pattern Automated: Computer-Assisted,Image Computer-Assisted: Enhancement,Image Enhancement: Intelligence,Computer Interpretation, Recognition, Results,Sensitivity Retrieval,Information Retrieval: Simulation,Data Specificity Statistical,Image Statistical,Pattern Storage and methods,Image methods,Information methods,Models, methods,Reproducibility of
%N 6
%P 929--44
%R 10.1109/TPAMI.2007.1046
%T Toward objective evaluation of image segmentation algorithms.
%U http://www.ncbi.nlm.nih.gov/pubmed/17431294
%V 29
%X Unsupervised image segmentation is an important component in many
image understanding algorithms and practical vision systems. However,
evaluation of segmentation algorithms thus far has been largely subjective,
leaving a system designer to judge the effectiveness of a technique
based only on intuition and results in the form of a few example
segmented images. This is largely due to image segmentation being
an ill-defined problem-there is no unique ground-truth segmentation
of an image against which the output of an algorithm may be compared.
This paper demonstrates how a recently proposed measure of similarity,
the Normalized Probabilistic Rand (NPR) index, can be used to perform
a quantitative comparison between image segmentation algorithms using
a hand-labeled set of ground-truth segmentations. We show that the
measure allows principled comparisons between segmentations created
by different algorithms, as well as segmentations on different images.
We outline a procedure for algorithm evaluation through an example
evaluation of some familiar algorithms-the mean-shift-based algorithm,
an efficient graph-based segmentation algorithm, a hybrid algorithm
that combines the strengths of both methods, and expectation maximization.
Results are presented on the 300 images in the publicly available
Berkeley Segmentation Data Set.
@article{Unnikrishnan2007,
abstract = {Unsupervised image segmentation is an important component in many
image understanding algorithms and practical vision systems. However,
evaluation of segmentation algorithms thus far has been largely subjective,
leaving a system designer to judge the effectiveness of a technique
based only on intuition and results in the form of a few example
segmented images. This is largely due to image segmentation being
an ill-defined problem-there is no unique ground-truth segmentation
of an image against which the output of an algorithm may be compared.
This paper demonstrates how a recently proposed measure of similarity,
the Normalized Probabilistic Rand (NPR) index, can be used to perform
a quantitative comparison between image segmentation algorithms using
a hand-labeled set of ground-truth segmentations. We show that the
measure allows principled comparisons between segmentations created
by different algorithms, as well as segmentations on different images.
We outline a procedure for algorithm evaluation through an example
evaluation of some familiar algorithms-the mean-shift-based algorithm,
an efficient graph-based segmentation algorithm, a hybrid algorithm
that combines the strengths of both methods, and expectation maximization.
Results are presented on the 300 images in the publicly available
Berkeley Segmentation Data Set.},
added-at = {2013-09-29T14:16:50.000+0200},
author = {Unnikrishnan, Ranjith and Pantofaru, Caroline and Hebert, Martial},
biburl = {https://www.bibsonomy.org/bibtex/2ab3b0b2a3f3f0508f8c339817b90df70/guillem.palou},
doi = {10.1109/TPAMI.2007.1046},
interhash = {0dcdb79af699ec29fa9982143f3faca9},
intrahash = {ab3b0b2a3f3f0508f8c339817b90df70},
issn = {0162-8828},
journal = {IEEE TPAMI},
keywords = {Algorithms,Artificial Automated,Pattern Automated: Computer-Assisted,Image Computer-Assisted: Enhancement,Image Enhancement: Intelligence,Computer Interpretation, Recognition, Results,Sensitivity Retrieval,Information Retrieval: Simulation,Data Specificity Statistical,Image Statistical,Pattern Storage and methods,Image methods,Information methods,Models, methods,Reproducibility of},
number = 6,
pages = {929--44},
pmid = {17431294},
timestamp = {2013-09-29T14:16:50.000+0200},
title = {{Toward objective evaluation of image segmentation algorithms.}},
url = {http://www.ncbi.nlm.nih.gov/pubmed/17431294},
volume = 29,
year = 2007
}