ICCVG 2010
tuesday, 22 January 2019
switch to english

Sign in

Registered user

Login

Password

Home > Description

Invited Lectures at the Conference ICCVG 2010

The invited lectures will be given by:

Professor Silvana Dellepiane

http://cucciolo.dibe.unige.it/NUMIP/

Lecture title:
Fuzzy segmentation based on intensity-connectedness

Summary:

In this lecture a multi-region segmentation method based on fuzzy intensity-connectedness is described.

The first proposed seed-segmentation method is the original work of Seeded-Region-Growing (SRG) introduced by Adams and Bishop [1]. Despite its simplicity, SRG results are very good, but it was often pointed out that they are dependent on the order of analysis. To this end, some solutions have been proposed [2]. In all these works no fuzzy measure neither fuzzy processing are applied. The first segmentation algorithms based on fuzzy connectedness were independently proposed by Dellepiane et al., introducing in [4] the intensity-connectedness concept, and by Udupa et al. who defined the local fuzzy relation called "affinity" in [5].

In the present lecture, the adaptive growing mechanism, originally proposed in [3,4] for fuzzy intensity-connectedness measurement, is extended to the multi-seed case, to the volumetric third dimension, and to the multi-temporal analysis. Such a growing mechanism, starting from a small number of seeds, is adaptive to the actual data content and is able to correctly take into account local and global connectedness relationships. It can be proved that this growing mechanism assures the best path selection and turns to be completely independent of the order of analysis. This allows to be sure that the maximum membership decision step required for the extraction of fuzzy connectedness is not affected by any polarization or error. Such an aspect is mainly critical when analyzing real images, where contrast is poor and blurred, and spatial inhomogeneities are present, in addition to noise.

The obtained result is a fuzzy segmentation where a membership value is associated to each analyzed spel (i.e., a pixel in the 2D space or a voxel in the 3D space, respectively) related to each seed-class. The de-fuzzification gives rise to a hard result, by applying the maximum membership criterion as usually applied in fuzzy cluster analysis.

At the same time, the uncertainty degree associated with the classes remains available for each analyzed spel. In such a way it is possible to discard doubtful assignments, when one should prefer not to decide instead of taking a wrong decision. The application of the method allows to focus the attention to some regions of interest (ROI), avoiding the global segmentation of the whole image when it is not required. The handling of uncertainty allows to avoid the use of any parameter.

Even though the method can be applied to any kind of digital image or digital volume, performance evaluation is presented in the biomedical domain, referring to the results obtained from two standard image databases from MRI, the former made of synthetic brain volumes, the latter made of real brain volumes. A quantitative performance evaluation is carried out on the two data sets where sensitivity, specificity, and accuracy for the segmentation of white matter, gray matter and cerebrospinal fluid are computed.

References

  1. Adams, R., Bischof, L.: Seeded Region Growing. In: IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 16-6, pp. 641-647 (1994)
  2. Mehnert, A., Jackway, P.: An Improved Seeded Region Growing Algorithm. In: Pattern Recognition Letters, vol.18, pp.106-1071 (1997)
  3. Dellepiane, S., Fontana, F., Vernazza, G.: Nonlinear Image Labelling for Multivalued Segmentation. In: IEEE Trans. on Image Processing, vol. 3, pp. 429-446 (1996)
  4. Dellepiane, S., Fontana, F.: Extraction of Intensity Connectedness for Image Processing. In: Pattern Recognition Letters, vol.16, pp.313-324 (1995)
  5. Udupa, J.K, Samarasekera, S.: Fuzzy Connectedness and Object Definition: Theory, Algorithms, and Applications in Image Segmentation. In Graphical Models and Image Processing, vol. 58, pp. 246-261, (1996).
  Professor André Gagalowicz

http://www-roc.inria.fr/mirages/

Lecture title:
3D model-based face tracking including non verbal expressions

Summary:

We propose a new technique for 3D face tracking in video sequences, which works without markers and once initialized, doesn't require any further interaction. A specific geometric model of the face (in a neutral position, for example) is used as input as well as a realistic face animation model. The algorithm includes a new method for precise face expression tracking in a video sequence which uses this hierarchical animation system built over a morphable polygonal 3D face model. Its low-level animation mechanism is based upon MPEG-4 specification which is implemented via local point-driven mesh deformations adaptive to the face geometry. The set of MPEG-4 animation parameters is in its turn controlled by a higher-level system based upon facial muscles structure. That allows us to perform precise tracking of complicated facial expressions as well as to produce face-to-face retargeting by transmitting the expression parameters to the different faces.

  Professor Ryszard Tadeusiewicz

http://www.uci.agh.edu.pl/english/tad/

Lecture title:
New challenge in Machine Vision - Automatic understanding of images

Ryszard Tadeusiewicz, Marek R. Ogiela

Summary:

Machine vision is currently well developed area of computer science with many renown scientific achievements and with numerous valuable practical applications. If we take into account typical set topics and consecutive steps belonging to machine vision we can distinguish frame acquisition, image processing, picture analysis and pattern recognition. Almost every step in such intelligent image analysis is now developed wry well. Although permanent increasing number and usefulness of algorithms for image data processing - general and dedicated for particular applications - we can assess collection of existing methods as rich enough and developed very well. Image data analysis and interpretation is also good developed area of scientific knowledge, with most important technologies discovered yet and permanently optimized. The same remarks can be addressed to pattern recognition and automatic classification methods, because such problems was studied by lot of researchers since over fifty years.

Therefore some specialists formulated opinion, that typical image processing, analysis, and recognition are now domains closed from the scientific point of view, and only necessary area of professional activity is development of new applications as well as building useful computer programs for machine vision users. In fact for many purposes present computer vision systems are enough efficient. Moreover in numerous applications computer image processing systems can be faster and more precise than human eye.

Nevertheless machine vision as a scientific research area is still open, because exist big gap between possibilities of even best computer vision systems and even average visual competences of the human mind. This gap is evident in almost all non-trivial application of artificial image analysis devices. The merit of this gap is related to fundamental difference between image processing and recognition (which can be performed automatically) and image content understanding, which must be elaborated by intelligent human mind.

Understanding of the images is the key feature which is inherent to human vision system and is absolutely not available in current computer vision system. Every human perception is definitely knowledge based. We can use visual information only when we can understand it. If the image is not intelligible for somebody - it is worthless for any application. We cannot use it and usually we cannot also precisely remember this image - if the subject and merit sense of the image is mysterious or obscure.

Therefore if we try replace human eye (and mind!) by the automatic system in some advanced visual tasks - we must cross this gap. Moreover for further development of intelligent image analysis systems automatic understanding of images merit content is absolutely necessary. In this invited lecture we explain, what does it means "understanding of the images" both for human being and for computers. Next we propose methods which can be used for semantic description of the image merit content. We point out in this context the role of mathematical linguistic and we discuss graph grammar based artificial languages used for image semantic expression. Next we introduce and discuss cognitive resonance method, which is developed by us for automatic understanding of the images and which help us achieve many successes in practical cognitive computer vision systems applications. At the end of lecture we try sketch general structure of the automatic image understanding system and we propose theoretical solutions for fundamental problems related to the automatic understanding of the images.

back           
Asociation for Image Processing Polish-Japanese Institute of Information Technology Springer, Lecture Notes in Computer Science Faculty of Applied Informatics and Mathematics, Warsaw University of Life Sciences

Association for Image Processing

Faculty of Applied Informatics and Mathematics

Photo of Warsaw by www.zdjeciawarszawy.pl

v. 2012.2.1.1 eConf © 2008-2016 Piotr Ku¼niacki

Last modification 14-09-2010

Site counter 26488 Online users 4