The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Dans cet article, nous présentons les schémas suivants pour une recherche d'images basée sur le contenu : (1) Un algorithme de recherche d'images rapide qui peut réduire considérablement le calcul de similarité par rapport à une comparaison complète de chaque image de base de données. (2) Un schéma de représentation d'image compact qui peut décrire les informations globales/locales des images et fournir des performances de récupération réussies. Pour des recherches rapides, un arbre est construit en divisant avec succès les nœuds selon le niveau de profondeur souhaité en travaillant de la racine aux nœuds feuilles à l'aide de l'algorithme k-means. Lorsque la requête est terminée, nous parcourons l'arborescence de haut en bas en minimisant l'itinéraire emprunté entre l'image de la requête et le centroïde du nœud jusqu'à ce que nous rencontrions les nœuds non divisés. Au sein des nœuds indivis, l'algorithme d'inégalité triangulaire est utilisé pour trouver les images les plus similaires à la requête. Pour une représentation d'image compacte, les caractéristiques de l'histogramme couleur RVB qui sont quantifiées dans 16 bacs pour chacun des canaux R, V et B sont utilisées pour les informations globales. La teinte, la saturation et la valeur dominantes qui sont extraites de l'histogramme conjoint HSV dans les régions localisées de l'image sont utilisées pour les informations locales. Ces fonctionnalités sont suffisamment compactes pour indexer les fonctionnalités d’images dans de grands systèmes de bases de données. Pour les expériences sur l'efficacité de la récupération, l'utilisation de la méthode proposée a apporté des avantages substantiels en termes de performances en réduisant le calcul de similarité d'image jusqu'à une moyenne de 96 % et pour les expériences sur l'efficacité de la récupération, dans le meilleur des cas, elle fournit un rappel de 36.8 %. taux pour une image de requête de baleine et un taux de précision de 100 % pour une image de requête d'aigle. La performance globale était un taux de rappel de 20.0 % et un taux de précision de 72.5 %.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Hun-Woo YOO, Dong-Sik JANG, Yoon-Kyoon NA, "An Efficient Indexing Structure and Image Representation for Content-Based Image Retrieval" in IEICE TRANSACTIONS on Information,
vol. E85-D, no. 9, pp. 1390-1398, September 2002, doi: .
Abstract: In this paper, we present the following schemes for a content-based image search: (1) A fast image search algorithm that can significantly reduce similarity calculation compared to a full comparison of every database image. (2) A compact image representation scheme that can describe the global/local information of the images and provide successful retrieval performance. For fast searches, a tree is constructed by successfully dividing nodes into the desired depth level by working from the root to the leaf nodes using the k-means algorithm. When the query is completed, we traverse the tree top-down by minimizing the route taken between the query image and node centroid until we meet the undivided nodes. Within undivided nodes, the algorithm of triangle inequality is used to find the images most similar to the query. For compact image representation, RGB color histogram features which are quantized into 16 bins each of the R, G, and B channels are used for global information. Dominant hue, saturation, and value which are extracted from the HSV joint histogram in the localized regions within the image are used for local information. These features are sufficiently compact to index image features in large database systems. For experiments on the retrieval efficiency, the use of the proposed method provided substantial performance benefits by reducing the image similarity calculation up to an average of a 96% and for experiments on the retrieval effectiveness, in the best case, it provide a 36.8% recall rate for a whale query image and a 100% precision rate for an eagle query image. The overall performance was a 20.0% recall rate and a 72.5% precision rate.
URL: https://global.ieice.org/en_transactions/information/10.1587/e85-d_9_1390/_p
Copier
@ARTICLE{e85-d_9_1390,
author={Hun-Woo YOO, Dong-Sik JANG, Yoon-Kyoon NA, },
journal={IEICE TRANSACTIONS on Information},
title={An Efficient Indexing Structure and Image Representation for Content-Based Image Retrieval},
year={2002},
volume={E85-D},
number={9},
pages={1390-1398},
abstract={In this paper, we present the following schemes for a content-based image search: (1) A fast image search algorithm that can significantly reduce similarity calculation compared to a full comparison of every database image. (2) A compact image representation scheme that can describe the global/local information of the images and provide successful retrieval performance. For fast searches, a tree is constructed by successfully dividing nodes into the desired depth level by working from the root to the leaf nodes using the k-means algorithm. When the query is completed, we traverse the tree top-down by minimizing the route taken between the query image and node centroid until we meet the undivided nodes. Within undivided nodes, the algorithm of triangle inequality is used to find the images most similar to the query. For compact image representation, RGB color histogram features which are quantized into 16 bins each of the R, G, and B channels are used for global information. Dominant hue, saturation, and value which are extracted from the HSV joint histogram in the localized regions within the image are used for local information. These features are sufficiently compact to index image features in large database systems. For experiments on the retrieval efficiency, the use of the proposed method provided substantial performance benefits by reducing the image similarity calculation up to an average of a 96% and for experiments on the retrieval effectiveness, in the best case, it provide a 36.8% recall rate for a whale query image and a 100% precision rate for an eagle query image. The overall performance was a 20.0% recall rate and a 72.5% precision rate.},
keywords={},
doi={},
ISSN={},
month={September},}
Copier
TY - JOUR
TI - An Efficient Indexing Structure and Image Representation for Content-Based Image Retrieval
T2 - IEICE TRANSACTIONS on Information
SP - 1390
EP - 1398
AU - Hun-Woo YOO
AU - Dong-Sik JANG
AU - Yoon-Kyoon NA
PY - 2002
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E85-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2002
AB - In this paper, we present the following schemes for a content-based image search: (1) A fast image search algorithm that can significantly reduce similarity calculation compared to a full comparison of every database image. (2) A compact image representation scheme that can describe the global/local information of the images and provide successful retrieval performance. For fast searches, a tree is constructed by successfully dividing nodes into the desired depth level by working from the root to the leaf nodes using the k-means algorithm. When the query is completed, we traverse the tree top-down by minimizing the route taken between the query image and node centroid until we meet the undivided nodes. Within undivided nodes, the algorithm of triangle inequality is used to find the images most similar to the query. For compact image representation, RGB color histogram features which are quantized into 16 bins each of the R, G, and B channels are used for global information. Dominant hue, saturation, and value which are extracted from the HSV joint histogram in the localized regions within the image are used for local information. These features are sufficiently compact to index image features in large database systems. For experiments on the retrieval efficiency, the use of the proposed method provided substantial performance benefits by reducing the image similarity calculation up to an average of a 96% and for experiments on the retrieval effectiveness, in the best case, it provide a 36.8% recall rate for a whale query image and a 100% precision rate for an eagle query image. The overall performance was a 20.0% recall rate and a 72.5% precision rate.
ER -