Hierarchy in picture segmentation: a stepwise optimization approach

Beaulieu, Jean-Marie; Goldberg, M.

doi:10.1109/34.16711

Cited by 239 publications

(118 citation statements)

References 18 publications

Supporting

Mentioning

115

Contrasting

Unclassified

Order By: Relevance

“…the difference of their average intensity) [1,2,3,4]. This bottom-up approach fits very well with the concept of irregular pyramids [5,6], and the main direction of this work is to show how the Active Paintbrush -an interactive segmentation tool developed for medical imaging [2] -and an automatic region merging [7,3,2] can be formulated based on the concepts of irregular pyramids and contraction kernels. This serves three goals: a) delivering a useful, practical application of contraction kernels, b) basing the description of segmentation methods on well-known concepts instead of their own, specialized representation, and c) demonstrating how a common representation facilitates the development of a more efficient integration of the above automatic and interactive methods.…”

Section: Introductionmentioning

confidence: 88%

“…Continuous Pyramids In fact, we can build "continuous pyramids" in which only one region is merged in every step, as done by the stepwise optimization used for the Active Paintbrush preprocessing [7,2]. In our context, the reduction factor between successive levels can be declared irrelevant:…”

Section: Interactive Navigation Of Continuous Pyramidsmentioning

confidence: 99%

“…Merge Tree Representation For this work, it is important to highlight the internal representation built within the first step, in which the automatic region merging interactively merges the two regions rated most similar (an equivalent approach is used in [7,3,2,8]). This process is continued until the whole image is represented by one big region, and at the same time a hierarchical description of the image is built up: a tree of merged regions, the leaves of which are the primitive regions of the initial oversegmentation (illustrated in Fig.…”

Section: The Active Paintbrush Toolmentioning

confidence: 99%

See 2 more Smart Citations

Annotated Contraction Kernels for Interactive Image Segmentation

Meine

2009

Graph-Based Representations in Pattern Recognition

View full text Add to dashboard Cite

Abstract. This article shows how the interactive segmentation tool termed "Active Paintbrush" and a fully automatic region merging can both be based on the theoretical framework of contraction kernels within irregular pyramids instead of their own, specialized data structures. We introduce "continous pyramids" in which we purposely drop the common requirement of a fixed reduction factor between successive levels, and we show how contraction kernels can be annotated for a fast navigation of such pyramids. Finally, we use these concepts for improving the integration of the automatic region merging and the interactive tool.

show abstract

Section: Introductionmentioning

confidence: 88%

Section: Interactive Navigation Of Continuous Pyramidsmentioning

confidence: 99%

Section: The Active Paintbrush Toolmentioning

confidence: 99%

See 1 more Smart Citation

Annotated Contraction Kernels for Interactive Image Segmentation

Meine

2009

Graph-Based Representations in Pattern Recognition

View full text Add to dashboard Cite

show abstract

“…Facing an absence of contextual knowledge, the only alternative which can enrich our knowledge concerning the signi"cance of our segmented groups is the creation of a hierarchy guided by the knowledge which emerges from the super"cial and deep image structure [9]. Furthermore, a hierarchy can be constructed either on a pixel-to-pixel [1,10] or on a region-to-region basis [2,11]. In the present work, we deal with information which can be retrieved from the super-"cial structure of the image and we strive towards the study of constructing hierarchical methods with respect to a region-based approach, wherein the initial pixel grouping is guided by the principles of the watershed analysis [2].…”

Section: Introductionmentioning

confidence: 99%

Low level image partitioning guided by the gradient watershed hierarchy

1999

View full text Add to dashboard Cite

In this paper, we study the construction of hierarchies for gradient watersheds. The underlying analysis is based on the fact that the gradient image provides information with respect to local discontinuity and similarity. Two algorithms which utilize this type of information are analyzed. The "rst one is an extended version of the dynamics of contours and the second is the algorithm which considers the minimum cost path saliency between two adjacent regional minima. For both algorithms, a stopping criterion which de"nes and extracts automatically each hierarchical level is proposed. An important hint to determine which one of the two algorithms has a superior behavior is the domination of the pro"le type of the surface (convex or concave) formed by the gradient image. Tests are discussed on arti"cially generated and real images. The proposed bottom-up procedure provides an essential aid to the precise classi"cation of anatomical objects when used together with a user-interface environment.1999 Elsevier Science B.V. All rights reserved. ZusammenfassungIm folgenden Beitrag werden wir die Konstruktion von Hierarchien fuK r Wendepunkte untersuchen. Die zugrundeliegende Analyse beruht auf da{ Gradientenbilder Informationen uK ber lokale DiskontinuitaK ten und A G hnlichkeiten beinhalten. Es werden zwei Algorithmen untersucht, die diese Art von Information nutzen. Der erste ist eine erweiterte Version der Dynamik von Konturen und der zweite Algorithmus betrachtet minimale SpruK nge der Kostenfunktion zwischen zwei benachbarten regionalen Minima. FuK r beide Algorithmen wird ein Abbruchkriterium vorgestellt, das automatisch jede hierarchische Stufe de"niert und extrahiert. Ein wichtiger Hinweis dafuK r, welcher der beiden Algorithmen ein besseres Verhalten liefert, ist die Dominanz des Pro"ltyps der Ober#aK che (konvex oder konkav), gegeben durch das Gradientenbild. Tests mit kuK nstlichen und realen Bildern werden diskutiert. Die vorgestellte bottom-up Methode liefert eine entscheidende Hilfe zur praK zisen Klassi"zierung anatomischer Objekte, wenn sie im Zusammenhang mit einer Benutzerschnittstelle eingesetzt wird.1999 Elsevier Science B.V. All rights reserved. Re2 sume2Dans cet article, nous eH tudions la construction de hieH rarchies pour les lignes de partage des eaux du gradient. L'analyse sous-jacente repose sur le fait que l'image gradient fournit de l'information relative aux discontinuiteH s et aux similariteH s 0165-1684/99/$ -see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 -1 6 8 4 ( 9 8 ) 0 0 2 3 2 -1 locales. Deux algorithmes qui utilisent ce type d'information sont analyseH s. Le premier est une version eH tendue de la dynamique des contours et le deuxie`me est un algorithme qui conside`re le chemin de cou( t minimum entre deux minima reH gionaux adjacents. Pour les deux algorithmes, nous proposons un crite`re d'arre( t qui deH "nit et extrait automatiquement chaque niveau hieH rarchique. Une indication importante pour deH terminer lequel des deux algorithmes a un...

show abstract

“…Accordingly, energy minimization methods and iterative algorithms are not necessary to solve the optimization problem. We have modified a particular segmentation energy first introduced in a discrete setting by Beaulieu and Goldberg [2]. The original model tends to obtain a partition with a few number of regions and small variances without a priori knowledge on the image.…”

Section: Introductionmentioning

confidence: 99%