Region-Based Image Retrieval Using an Object Ontology and Relevance Feedback
<p/> <p>An image retrieval methodology suited for search in large collections of heterogeneous images is presented. The proposed approach employs a fully unsupervised segmentation algorithm to divide images into regions and endow the indexing and retrieval system with content-based funct...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2004-01-01
|
| Series: | EURASIP Journal on Advances in Signal Processing |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1155/S1110865704401188 |
| _version_ | 1819071200245055488 |
|---|---|
| author | Kompatsiaris Ioannis Mezaris Vasileios Strintzis Michael G |
| author_facet | Kompatsiaris Ioannis Mezaris Vasileios Strintzis Michael G |
| author_sort | Kompatsiaris Ioannis |
| collection | DOAJ |
| description | <p/> <p>An image retrieval methodology suited for search in large collections of heterogeneous images is presented. The proposed approach employs a fully unsupervised segmentation algorithm to divide images into regions and endow the indexing and retrieval system with content-based functionalities. Low-level descriptors for the color, position, size, and shape of each region are subsequently extracted. These arithmetic descriptors are automatically associated with appropriate qualitative intermediate-level descriptors, which form a simple vocabulary termed <it>object ontology</it>. The object ontology is used to allow the qualitative definition of the high-level concepts the user queries for (<it>semantic objects</it>, each represented by a <it>keyword</it>) and their relations in a human-centered fashion. When querying for a specific semantic object (or objects), the intermediate-level descriptor values associated with both the semantic object and all image regions in the collection are initially compared, resulting in the rejection of most image regions as irrelevant. Following that, a relevance feedback mechanism, based on support vector machines and using the low-level descriptors, is invoked to rank the remaining potentially relevant image regions and produce the final query results. Experimental results and comparisons demonstrate, in practice, the effectiveness of our approach.</p> |
| first_indexed | 2024-12-21T17:18:03Z |
| format | Article |
| id | doaj.art-3a8368b4aad24515911d5e56092c4bdc |
| institution | Directory Open Access Journal |
| issn | 1687-6172 1687-6180 |
| language | English |
| last_indexed | 2024-12-21T17:18:03Z |
| publishDate | 2004-01-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | EURASIP Journal on Advances in Signal Processing |
| spelling | doaj.art-3a8368b4aad24515911d5e56092c4bdc2022-12-21T18:56:15ZengSpringerOpenEURASIP Journal on Advances in Signal Processing1687-61721687-61802004-01-0120046231946Region-Based Image Retrieval Using an Object Ontology and Relevance FeedbackKompatsiaris IoannisMezaris VasileiosStrintzis Michael G<p/> <p>An image retrieval methodology suited for search in large collections of heterogeneous images is presented. The proposed approach employs a fully unsupervised segmentation algorithm to divide images into regions and endow the indexing and retrieval system with content-based functionalities. Low-level descriptors for the color, position, size, and shape of each region are subsequently extracted. These arithmetic descriptors are automatically associated with appropriate qualitative intermediate-level descriptors, which form a simple vocabulary termed <it>object ontology</it>. The object ontology is used to allow the qualitative definition of the high-level concepts the user queries for (<it>semantic objects</it>, each represented by a <it>keyword</it>) and their relations in a human-centered fashion. When querying for a specific semantic object (or objects), the intermediate-level descriptor values associated with both the semantic object and all image regions in the collection are initially compared, resulting in the rejection of most image regions as irrelevant. Following that, a relevance feedback mechanism, based on support vector machines and using the low-level descriptors, is invoked to rank the remaining potentially relevant image regions and produce the final query results. Experimental results and comparisons demonstrate, in practice, the effectiveness of our approach.</p>http://dx.doi.org/10.1155/S1110865704401188image retrievalimage databasesimage segmentationontologyrelevance feedbacksupport vector machines |
| spellingShingle | Kompatsiaris Ioannis Mezaris Vasileios Strintzis Michael G Region-Based Image Retrieval Using an Object Ontology and Relevance Feedback EURASIP Journal on Advances in Signal Processing image retrieval image databases image segmentation ontology relevance feedback support vector machines |
| title | Region-Based Image Retrieval Using an Object Ontology and Relevance Feedback |
| title_full | Region-Based Image Retrieval Using an Object Ontology and Relevance Feedback |
| title_fullStr | Region-Based Image Retrieval Using an Object Ontology and Relevance Feedback |
| title_full_unstemmed | Region-Based Image Retrieval Using an Object Ontology and Relevance Feedback |
| title_short | Region-Based Image Retrieval Using an Object Ontology and Relevance Feedback |
| title_sort | region based image retrieval using an object ontology and relevance feedback |
| topic | image retrieval image databases image segmentation ontology relevance feedback support vector machines |
| url | http://dx.doi.org/10.1155/S1110865704401188 |
| work_keys_str_mv | AT kompatsiarisioannis regionbasedimageretrievalusinganobjectontologyandrelevancefeedback AT mezarisvasileios regionbasedimageretrievalusinganobjectontologyandrelevancefeedback AT strintzismichaelg regionbasedimageretrievalusinganobjectontologyandrelevancefeedback |