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 des études antérieures, la précision de la récupération de grandes bases de données d'images a été améliorée grâce à la réduction du fossé sémantique en combinant l'esquisse d'entrée avec un retour de pertinence. Une amélioration supplémentaire de la précision de la récupération est attendue en combinant chaque trait et son ordre de l'esquisse d'entrée avec le retour d'information sur la pertinence. Cependant, cela pose problème : l'effet du retour de pertinence dépend essentiellement de l'ordre des traits dans l'esquisse d'entrée. Bien qu’il soit théoriquement possible de considérer tous les ordres de traits possibles, cela poserait un problème réaliste de création d’une énorme quantité de données. Par conséquent, la technique présentée dans cet article vise à améliorer l'efficacité de la récupération en utilisant efficacement le retour de pertinence au moyen de l'exploration de données du croquis en tenant compte de la similarité dans l'ordre des traits. Pour vérifier l'efficacité de cette technique, une expérience de récupération a été menée à partir de 20,000 XNUMX images d'une collection, la Corel Photo Gallery, et l'expérience a pu confirmer une amélioration de l'efficacité de la récupération.
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
Takashi HISAMORI, Toru ARIKAWA, Gosuke OHASHI, "Query-by-Sketch Image Retrieval Using Similarity in Stroke Order" in IEICE TRANSACTIONS on Information,
vol. E93-D, no. 6, pp. 1459-1469, June 2010, doi: 10.1587/transinf.E93.D.1459.
Abstract: In previous studies, the retrieval accuracy of large image databases has been improved as a result of reducing the semantic gap by combining the input sketch with relevance feedback. A further improvement of retrieval accuracy is expected by combining each stroke, and its order, of the input sketch with the relevance feedback. However, this leaves as a problem the fact that the effect of the relevance feedback substantially depends on the stroke order in the input sketch. Although it is theoretically possible to consider all the possible stroke orders, that would cause a realistic problem of creating an enormous amount of data. Consequently, the technique introduced in this paper intends to improve retrieval efficiency by effectively using the relevance feedback by means of conducting data mining of the sketch considering the similarity in the order of strokes. To ascertain the effectiveness of this technique, a retrieval experiment was conducted using 20,000 images of a collection, the Corel Photo Gallery, and the experiment was able to confirm an improvement in the retrieval efficiency.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E93.D.1459/_p
Copier
@ARTICLE{e93-d_6_1459,
author={Takashi HISAMORI, Toru ARIKAWA, Gosuke OHASHI, },
journal={IEICE TRANSACTIONS on Information},
title={Query-by-Sketch Image Retrieval Using Similarity in Stroke Order},
year={2010},
volume={E93-D},
number={6},
pages={1459-1469},
abstract={In previous studies, the retrieval accuracy of large image databases has been improved as a result of reducing the semantic gap by combining the input sketch with relevance feedback. A further improvement of retrieval accuracy is expected by combining each stroke, and its order, of the input sketch with the relevance feedback. However, this leaves as a problem the fact that the effect of the relevance feedback substantially depends on the stroke order in the input sketch. Although it is theoretically possible to consider all the possible stroke orders, that would cause a realistic problem of creating an enormous amount of data. Consequently, the technique introduced in this paper intends to improve retrieval efficiency by effectively using the relevance feedback by means of conducting data mining of the sketch considering the similarity in the order of strokes. To ascertain the effectiveness of this technique, a retrieval experiment was conducted using 20,000 images of a collection, the Corel Photo Gallery, and the experiment was able to confirm an improvement in the retrieval efficiency.},
keywords={},
doi={10.1587/transinf.E93.D.1459},
ISSN={1745-1361},
month={June},}
Copier
TY - JOUR
TI - Query-by-Sketch Image Retrieval Using Similarity in Stroke Order
T2 - IEICE TRANSACTIONS on Information
SP - 1459
EP - 1469
AU - Takashi HISAMORI
AU - Toru ARIKAWA
AU - Gosuke OHASHI
PY - 2010
DO - 10.1587/transinf.E93.D.1459
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E93-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2010
AB - In previous studies, the retrieval accuracy of large image databases has been improved as a result of reducing the semantic gap by combining the input sketch with relevance feedback. A further improvement of retrieval accuracy is expected by combining each stroke, and its order, of the input sketch with the relevance feedback. However, this leaves as a problem the fact that the effect of the relevance feedback substantially depends on the stroke order in the input sketch. Although it is theoretically possible to consider all the possible stroke orders, that would cause a realistic problem of creating an enormous amount of data. Consequently, the technique introduced in this paper intends to improve retrieval efficiency by effectively using the relevance feedback by means of conducting data mining of the sketch considering the similarity in the order of strokes. To ascertain the effectiveness of this technique, a retrieval experiment was conducted using 20,000 images of a collection, the Corel Photo Gallery, and the experiment was able to confirm an improvement in the retrieval efficiency.
ER -