Other Free Encyclopedias » Online Encyclopedia » Encyclopedia - Featured Articles » Contributed Topics from K-O

Knowledge-Assisted Image and Video Analysis

domain segmentation objects prior

Definition: Knowledge-assisted image and video analysis refers to the use of domain-specific prior knowledge about images or video.

Segmentation of images and video is generally an ill-posed problem, i.e. for a given natural image or image sequence, there exists no unique solution to the segmentation problem. In order for the segmentation problem to be transformed to a well-posed one, restriction of the number of admissible solutions is necessary. To this end, the introduction of prior knowledge to the segmentation procedure, leading to the development of domain-specific knowledge-assisted analysis techniques, has been proposed.

Prior knowledge for a domain (e.g. Fl racing) typically includes the important objects that can be found in any given image or frame belonging to this domain (e.g. car, road, grass, sand, etc.), their characteristics (e.g. corresponding color models) and any relations between them. See example in Figure 1. Given this knowledge, there exists the well-posed problem of deciding, for each pixel, whether it belongs to any of the defined objects (and if so, to which one) or to none of them.

A knowledge-guided segmentation and labeling approach for still images is presented in , where an unsupervised fuzzy C-means clustering algorithm along with basic image processing techniques are used under the guidance of a knowledge base. The latter is constructed by automatically processing a set of ground-truth images to extract cluster-labeling rules.

An approach to the knowledge-assisted analysis of MPEG-2 video of the Fl racing domain is presented in, where the employed prior knowledge includes three semantic objects and their characteristics in the form of qualitative attributes (e.g. color homogeneity, indicating suitable pre-processing methods), numerical data generated via training and relations among semantic objects. This approach is extended in, where more semantic objects are specified and a genetic algorithm is applied to the atom regions initially generated via simple segmentation in order to find the optimal scene interpretation according to the domain conceptualization.


Knox, John (c. 1514–1572) - BIOGRAPHY, CRITICAL RECEPTION [next] [back] Knotts, Don

User Comments

Your email address will be altered so spam harvesting bots can't read it easily.
Hide my email completely instead?

Cancel or