|
| volgende >> |
Tijdschrift beschrijving
Alle jaargangen van het bijbehorende tijdschrift
Alle afleveringen van het bijbehorende jaargang
Alle artikelen van de bijbehorende aflevering
Details van artikel 1 van 12 gevonden artikelen
| Titel: |
A BPT Application: Semi-automatic Image Retrieval Tool |
| Auteur: |
Shirin Ghanbari John C. Woods Simon M. Lucas Hamid R. Rabiee |
| Verschenen in: |
International journal of computer science and information technology |
| Paginering: | Jaargang 2 (2010) nr. 1 pagina's 77-83 |
| Jaar: |
2010 |
| Inhoud: |
This work presents a semi-automatic tool forcontent retrieval. In contrast to traditional content basedimage retrieval systems that work with entire images, thetool we have developed handles individual objects. Theavailability of a pool of pre-segmented objects found usingregion analysis allows the human behaviour of presegmentationto be replicated. To generate defined objectsin the object pool, segmentation is performed using multidimensionalBinary Partition Trees (BPTs). The treestructure uses colour, spatial frequency edge histograms toform semantically meaningful tree nodes. The BPTs can beintuitively browsed and are stored within XML documentsfor ease of access and analysis. To find an object a nodefrom a query image is matched against the nodes of the BPTof the database image. These are matched according to acollection of MPEG-7 descriptors. Performance evaluationshows high quality segmentations and reliable retrievalresults. |
| Uitgever: |
Academy & Industry Research Collaboration Center (AIRCC) (provided by DOAJ) |
| Bronbestand: |
Elektronische Wetenschappelijke Tijdschriften |
Details van artikel 1 van 12 gevonden artikelen
| volgende >> |
Toegankelijkheidsverklaring