![](bg8.png)
approach or to the region level precision, segmenta-
tion process should follow a coarse to fine approach:
first the user can work with both approaches to obtain
a partition as close as possible to the desired partition,
using only the RAG (which implies efficient process-
ing time); then, using only markers (i.e., editing mark-
ers) over the PAG, he/she can refine the partition at a
pixel level precision.
5 CONCLUSIONS
We investigated how to perform transitions between
the hierarchical and the watershed from markers ap-
proaches. In particular, we discussed the effects ob-
served when these transitions involve changes in the
precision level of the partitions (pixels or regions).
These transitions have been implemented in conjunc-
tion with several interaction possibilities from both
watershed approaches as an interactive segmentation
tool. The implemented tool allows one to take advan-
tage of the best features of each approach and it is ex-
pected that the interaction effort necessary to produce
a desired result may be reduced.
An interesting characteristic of the proposed ap-
proach is that it allows producing partitions with pre-
cisions at pixel or region levels. As a consequence,
the tool may be used for different segmentation pur-
poses.
We plan some use case studies, in order to eval-
uate and improve the usability of the developed tool.
This study should take into consideration not only the
available functionalities but also aspects related to the
interaction process. For example, in applications in
which a precise segmentation is important, the inter-
action effort necessary to draw the markers can still
be very intense. It would be interesting if the tool
could provide means to easily place markers, on both
sides of the borders of interest. As an another exam-
ple, support for color image segmentation is currently
rather limited. These aspects, together with others to
be found in the use case studies, will guide our future
investigations.
ACKNOWLEDGEMENTS
This work is supported by CNPq, Brazil.
REFERENCES
Audigier, R. and Lotufo, R. A. (2007). Seed-relative seg-
mentation robustness of watershed and fuzzy connect-
edness approaches. In SIBGRAPI ’07: Proceedings
of the XX Brazilian Symposium on Computer Graph-
ics and Image Processing, pages 61–70, Washington,
DC, USA. IEEE Computer Society.
Audigier, R., Lotufo, R. A., and Couprie, M. (2005). The
tie-zone watershed: Definition, algorithm and applica-
tions. In Proceedings of IEEE International Confer-
ence on Image Processing (ICIP’05), pages 654–657.
Beucher, S. and Meyer, F. (1993). The morphological ap-
proach to segmentation: the watershed transforma-
tion. In Dougherty, E., editor, Mathematical morphol-
ogy in image processing, chapter 12, pages 433–481.
M. Dekker.
Falc
˜
ao, A. X., Stolfi, J., and Lotufo, R. A. (2004). The
image foresting transform: Theory, algorithms, and
applications. IEEE Trans. Pattern Anal. Mach. Intell.,
26(1):19–29.
Flores, F. C., Polidorio, A. M., and Lotufo, R. A. (2004).
Color image gradients for morphological segmenta-
tion: The weighted gradient improved by automatic
imposition of weights. In SIBGRAPI ’04: Proceed-
ings of the Computer Graphics and Image Processing,
XVII Brazilian Symposium, pages 146–153, Washing-
ton, DC, USA. IEEE Computer Society.
Hahn, H. K. and Peitgen, H.-O. (2003). IWT - interactive
watershed transform: A hierarchical method for effi-
cient interactive and automated segmentation of mul-
tidimensional gray-scale images. In Medical Imag-
ing 2003: Image Processing. Proceedings of the SPIE,
volume 5032, pages 643–653.
Lotufo, R. A. and Silva, W. (2002). Minimal set of markers
for the watershed. In Talbot, H. and Beare, R., ed-
itors, 6th International Symposium on Mathematical
Morphology, pages 359–368. CSIRO Publications.
Marcotegui, B. and Zanoguera, F. (2002). Image editing
tools based on multi-scale segmentation. In Talbot,
H. and Beare, R., editors, Proceedings of ISMM 2002,
pages 127–135.
Meyer, F. (1996). The dynamics of minima and contours. In
Maragos, P., Schafer, R. W., and Butt, M. A., editors,
Mathematical Morphology and its Applications to Im-
age and Signal Processing, pages 329–336. Kluwer
Academic Publishers.
Meyer, F. (2001a). Hierarchies of partitions and morpholog-
ical segmentation. In Scale-Space ’01: Proceedings
of the Third International Conference on Scale-Space
and Morphology in Computer Vision, pages 161–182,
London, UK. Springer-Verlag.
Meyer, F. (2001b). An overview of morphological segmen-
tation. International Journal of Pattern Recognition
and Artificial Intelligence, 15(7):1089–1118.
Zanoguera, F., Marcotegui, B., and Meyer, F. (1999). A
toolbox for interactive segmentation based on nested
partitions. In Proceedings of the International Con-
ference on Image Processing, volume 1, pages 21–25.
INTERACTIVE IMAGE SEGMENTATION WITH INTEGRATED USE OF THE MARKERS AND THE
HIERARCHICAL WATERSHED APPROACHES
193