Results 1  10
of
96
2D Euclidean distance transform algorithms: A comparative survey
 ACM COMPUTING SURVEYS
, 2008
"... The distance transform (DT) is a general operator forming the basis of many methods in computer vision and geometry, with great potential for practical applications. However, all the optimal algorithms for the computation of the exact Euclidean DT (EDT) were proposed only since the 1990s. In this wo ..."
Abstract

Cited by 63 (5 self)
 Add to MetaCart
The distance transform (DT) is a general operator forming the basis of many methods in computer vision and geometry, with great potential for practical applications. However, all the optimal algorithms for the computation of the exact Euclidean DT (EDT) were proposed only since the 1990s. In this work, stateoftheart sequential 2D EDT algorithms are reviewed and compared, in an effort to reach more solid conclusions regarding their differences in speed and their exactness. Six of the best algorithms were fully implemented and compared in practice.
A Graphbased Approach for Multiscale Shape Analysis
, 2003
"... This paper presents the advantages of computing two recently proposed shape descriptors, multiscale fractal dimension and contour saliences, using the image foresting transforma graphbased approach to the design of image processing operators. It introduces a robust approach to estimate contour ..."
Abstract

Cited by 48 (20 self)
 Add to MetaCart
This paper presents the advantages of computing two recently proposed shape descriptors, multiscale fractal dimension and contour saliences, using the image foresting transforma graphbased approach to the design of image processing operators. It introduces a robust approach to estimate contour saliences (peaks of high curvature) by exploiting the relation between contour and skeleton. The paper also compares both shape descriptors to fractal dimension, Fourier descriptors, and moment invariants with respect to their invariance to object characteristics that belong to a same class (compactability) and to their discriminatory ability to separate objects that belong to distinct classes (separability).
Power Watershed: A Unifying GraphBased Optimization Framework
, 2011
"... In this work, we extend a common framework for graphbased image segmentation that includes the graph cuts, random walker, and shortest path optimization algorithms. Viewing an image as a weighted graph, these algorithms can be expressed by means of a common energy function with differing choices of ..."
Abstract

Cited by 42 (8 self)
 Add to MetaCart
In this work, we extend a common framework for graphbased image segmentation that includes the graph cuts, random walker, and shortest path optimization algorithms. Viewing an image as a weighted graph, these algorithms can be expressed by means of a common energy function with differing choices of a parameter q acting as an exponent on the differences between neighboring nodes. Introducing a new parameter p that fixes a power for the edge weights allows us to also include the optimal spanning forest algorithm for watershed in this same framework. We then propose a new family of segmentation algorithms that fixes p to produce an optimal spanning forest but varies the power q beyond the usual watershed algorithm, which we term power watershed. In particular when q = 2, the power watershed leads to a multilabel, scale and contrast invariant, unique global optimum obtained in practice in quasilinear time. Placing the watershed algorithm in this energy minimization framework also opens new possibilities for using unary terms in traditional watershed segmentation and using watershed to optimize more general models of use in applications beyond image segmentation.
ContentBased Image Retrieval: Theory and Applications
 Revista de Informática Teórica e Aplicada
"... Advances in data storage and image acquisition technologies have enabled the creation of large image datasets. In this scenario, it is necessary to develop appropriate information systems to efficiently manage these collections. The commonest approaches use the socalled ContentBased Image Retrieva ..."
Abstract

Cited by 35 (18 self)
 Add to MetaCart
(Show Context)
Advances in data storage and image acquisition technologies have enabled the creation of large image datasets. In this scenario, it is necessary to develop appropriate information systems to efficiently manage these collections. The commonest approaches use the socalled ContentBased Image Retrieval (CBIR) systems. Basically, these systems try to retrieve images similar to a userdefined specification or pattern (e.g., shape sketch, image example). Their goal is to support image retrieval based on content properties (e.g., shape, color, texture), usually encoded into feature vectors. One of the main advantages of the CBIR approach is the possibility of an automatic retrieval process, instead of the traditional keywordbased approach, which usually requires very laborious and timeconsuming previous annotation of database images. The CBIR technology has been used in several applications such as fingerprint identification, biodiversity information systems, digital libraries, crime prevention, medicine, historical research, among others. This paper aims to introduce the problems and challenges concerned with the creation of CBIR systems, to describe the existing solutions and applications, and to present the state of the art of the existing research in this area.
Some links between mincuts, optimal spanning forests and watersheds
, 2007
"... Keywords: Di erent optimal structures: minimum cuts, minimum spanning forests and shortestpath forests, have been used as the basis for powerful image segmentation procedures. The wellknown notion of watershed also falls into this category. In this paper, we present some new results about the link ..."
Abstract

Cited by 24 (3 self)
 Add to MetaCart
(Show Context)
Keywords: Di erent optimal structures: minimum cuts, minimum spanning forests and shortestpath forests, have been used as the basis for powerful image segmentation procedures. The wellknown notion of watershed also falls into this category. In this paper, we present some new results about the links which exist between these di erent approaches. Especially, we show that mincuts coincide with watersheds for some particular weight functions. mincuts, spanning forests, watersheds, shortestpath forests.
IFTWatershed from GrayScale Marker
, 2002
"... The watershed transform and the morphological reconstruction are two of the most important operators for image segmentation in the framework of mathematical morphology. In many situations, the segmentation requires the classical watershed transform of a reconstructed image. In this paper, we intro ..."
Abstract

Cited by 17 (11 self)
 Add to MetaCart
The watershed transform and the morphological reconstruction are two of the most important operators for image segmentation in the framework of mathematical morphology. In many situations, the segmentation requires the classical watershed transform of a reconstructed image. In this paper, we introduce the IFTwatershed from gray scale marker  a method to compute at same time, the reconstruction and the classical watershed transform of the reconstructed image, without explicit computation of any regional minima. The method is based on the Image Foresting Transform (IFT)  a unified and efficient approach to reduce image processing problems to a minimumcost path forest problem in a graph. As additional contributions, we demonstrate that (i) the cost map of the IFTwatershed from markers is identical to the output of the superior gray scale reconstruction; (ii) other reconstruction algorithms are not watersheds; and (iii) the proposed method achieves competitive advantages as compared to the current classical watershed approach. 1
Supervised pattern classification based on optimumpath forest
 INTERN. JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY (IJIST
, 2009
"... We present a supervised classification method which represents each class by one or more optimumpath trees rooted at some key samples, called prototypes. The training samples are nodes of a complete graph, whose arcs are weighted by the distances between the feature vectors of their nodes. Prototyp ..."
Abstract

Cited by 17 (11 self)
 Add to MetaCart
(Show Context)
We present a supervised classification method which represents each class by one or more optimumpath trees rooted at some key samples, called prototypes. The training samples are nodes of a complete graph, whose arcs are weighted by the distances between the feature vectors of their nodes. Prototypes are identified in all classes and the minimization of a connectivity function by dynamic programming assigns to each training sample a minimumcost path from its most strongly connected prototype. This competition among prototypes partitions the graph into an optimumpath forest rooted at them. The class of the samples in an optimumpath tree is assumed to be the same of its root. A test sample is classified similarly, by identifying which tree would contain it, if the sample were part of the training set. By choice of the graph model and connectivity function, one can devise other optimumpath forest classifiers. We present one of them, which is fast, simple, multiclass, parameter independent, does not make any assumption about the shapes of the classes, and can handle some degree of overlapping between classes. We also propose a general algorithm to learn from errors on an evaluation set without increasing the training set, and show the advantages of our method with respect to SVM, ANNMLP, and kNN classifiers in several experiments with datasets of various types.
Contour Salience Descriptors for Effective Image Retrieval and Analysis
 IMAGE AND VISION COMPUTING
, 2007
"... This work exploits the resemblance between contentbased image retrieval and image analysis with respect to the design of image descriptors and their effectiveness. In this context, two shape descriptors are proposed: contour saliences and segment saliences. Contour saliences revisits its original d ..."
Abstract

Cited by 16 (10 self)
 Add to MetaCart
(Show Context)
This work exploits the resemblance between contentbased image retrieval and image analysis with respect to the design of image descriptors and their effectiveness. In this context, two shape descriptors are proposed: contour saliences and segment saliences. Contour saliences revisits its original definition, where the location of concave points was a problem, and provides a robust approach to incorporate concave saliences. Segment saliences introduces salience values for contour segments, making it possible to use an optimal matching algorithm as distance function. The proposed descriptors are compared with convex contour saliences, curvature scale space, and beam angle statistics using a fish database with 11,000 images organized in 1,100 distinct classes. The results indicate segment saliences as the most effective descriptor for this particular application and confirm the improvement of the contour salience descriptor in comparison with convex contour saliences.
Watershed by image foresting transform, tiezone, and theoretical relationships with other watershed definitions
, 2007
"... ..."
Image Segmentation by Tree Pruning
 IN PROC. OF THE XVII BRAZILLIAN SYMPOSIUM ON COMPUTER GRAPHICS AND IMAGE PROCESSING
, 2004
"... The Image Foresting Transform (IFT) has been proposed for the design of image operators based on connectivity. The IFT reduces image processing problems into a minimumcost path forest problem in a graph derived from the image. It has been successfully used for image filtering, segmentation, and anal ..."
Abstract

Cited by 14 (10 self)
 Add to MetaCart
The Image Foresting Transform (IFT) has been proposed for the design of image operators based on connectivity. The IFT reduces image processing problems into a minimumcost path forest problem in a graph derived from the image. It has been successfully used for image filtering, segmentation, and analysis. In this work, we propose a novel image operator which solves segmentation by pruning trees of the forest. First, an IFT is applied to create an optimumpath forest whose roots are pixels selected inside a desired object. In this forest, the background consists of a few subtrees rooted at pixels on the object's boundary. These boundary pixels are identified and their subtrees are eliminated, such that the remaining forest defines the object. The tree pruning is an effective alternative to situations where image segmentation methods based on competing seeds fail. We present an interactive implementation of the treepruning technique, show several examples and discuss some experiments toward fully automatic segmentation.