9:30, IMDSP-L5.1
ROBUST SEGMENTATION AND REPRESENTATION OF FOREGROUND KEY-REGIONS IN VIDEO SEQUENCES
J. RUIZ, P. SALEMBIER
This paper deals with the extraction and characterization of foreground objects (represented by key-regions) in video sequences. The algorithm first computes the mosaic image representing the background information and then extracts foreground key-regions. In this last step, the foreground key-regions are progressively extracted taking into account the reliability of the contour information. This extraction step is based on morphological tools. Finally, the foreground key-regions are characterized by their shape, texture and motion trajectory. Moreover, some information about the temporal evolution of non rigid objects is also extracted. This feature extraction algorithm is particularly suitable for the indexing, search and retrieval applications.

9:50, IMDSP-L5.2
REGULARIZED SHAPE DEFORMATION FOR IMAGE SEGMENTATION
S. WANG, Z. LIANG
This paper presents a new method for image segmentation by deforming the object shape in a template. The deformation process is controlled using a thin-plate spline kernel based regularization method. The proposed method is especially useful for 2D-based segmentation of 3D medical images by treating segmented slices as templates for their neighboring unsegmented slices. We have applied the proposed method to extract the scalp contours in brain cryosection images with very encouraging results.

10:10, IMDSP-L5.3
WAVELET BASED HALFTONE SEGMENTATION AND DESCREENING FILTER DESIGN
C. KUO, R. RAO, G. THOMPSON
The advent of electronic publication creates strong interest in converting existing printed documents into electronic formats. During this process, image reproduction problems can occur due to the formation of Moire patterns in the screened halftone areas. Therefore, optimal quality of a scanned document is achieved if halftone regions are first identified and processed separately. In this paper, we propose a complete algorithm to achieve this objective. A wavelet based halftone segmentation algorithm is first designed to locate possible halftone regions using a decision function. We then introduce a suboptimal FIR descreening filter to efficiently handle various screening frequencies and angles. Experimental results are offered to illustrate the performance of our algorithm.

10:30, IMDSP-L5.4
A GENERIC FUZZY RULE BASED TECHNIQUE FOR IMAGE SEGMENTATION
G. KARMAKAR, L. DOOLEY
Many fuzzy clustering based techniques do not incorporate spatial relationships of the pixels, while all fuzzy rule-based image segmentation techniques tend to be very much application dependent. In most techniques, the structure of the membership functions are predefined and their parameters are either automatically or manually determined. This paper addresses the aforementioned problems by introducing a general fuzzy rule based image segmentation technique, which is application independent and can also incorporate the spatial relationships of the pixels. It also proposes the automatic defining of the structure of the membership functions. A qualitative comparison is made between the segmentation results using this method and the popular fuzzy c-means (FCM) applied to two types of images: light intensity (LI) and X-ray of human vocal tract. The results clear show that this method exhibits significant improvements over FCM for both types of images.

10:50, IMDSP-L5.5
TEXTURE-SPACE SEGMENTATION AND MULTI-RESOLUTION MAPPING FOR FORESTRY APPLICATIONS
M. HILL, Y. CHANG, V. IYENGAR, C. LI
Forestry management requires careful and intensive planning efforts to ensure optimal yield, ecological stability, and regulatory compliance. In this paper, we describe a method of identifying wetlands and producing maps of their extent from commonly available, remotely-sensed imagery. This method provides a large labor savings over both field inspections and manual photo inspections. The enhanced accuracy translates into better timber harvest planning and better conservation of the wetlands.

11:10, IMDSP-L5.6
EXTRACTION OF SEMANTIC DESCRIPTION OF EVENTS USING BAYESIAN NETWORKS
A. EKIN, A. TEKALP, R. MEHROTRA
In this paper, we use Bayesian belief networks to statistically model the trends for event detection. We automatically detect non-rigid object trajectories for object motion units. Then, we use dominant and secondary trajectories of a single object in several consecutive motion units to understand semantic actions or those of more than one object to recognize semantic interactions between objects. We demonstrate sample Bayesian networks to detect events and extract the event descriptions, such as ``catch the ball'', ``throw the ball'' and ``walk''.