全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

利用颜色和熵提取感兴趣区域的感性图像检索

DOI: 10.11834/jig.20060481

Keywords: 感性图像检索,感兴趣区域,,神经网络

Full-Text   Cite this paper   Add to My Lib

Abstract:

感性图像检索是一种新型的检索技术,这种检索具有较高的复杂性。一幅图像中能够使人们产生感性认识的可能只有部分区域,准确地找到感兴趣区域能有效地降低复杂度。作为图像基本特征的颜色对人的感觉有重要的影响,颜色的差异和对比使人产生了不同的情感。同时,图像的熵也反应了图像中包含信息量的大小,图像的熵也是引起人们产生感性认识的一个度量。提出了利用图像的颜色和熵提取感兴趣区域进行感性图像检索的方法,通过BP神经网络将感兴趣区域的颜色特征和熵映射到情感特征空间,具有较好的检索效果。

References

[1]  Kuroda Kazuhiro,Masafumi Hagiwara.An image retrieval system by impression words and specific object names-IRIS[J].Neurocomputing,2002,43 (1):259 ~ 276.
[2]  The Color Science Association of Japan.Color Science Handbook[M].Tokyo,Japan:Tokyo University Press,1991.
[3]  Toshiyuki Iwaoka,Hideyuki Kobayahsi,shunji Ota.Digital safari guidebook with image retrieval[A].In:Proceedings of IEEE International Conference on Multimedia Computing and Systems[C],Centro Affari,Firenze,Italy,1999,2:1011 ~1012.
[4]  Takashi Hayashi,Masafumi Hagiwara.An image retrieval system to estimate impression words from images using a neural network[A].In:Proceedings of IEEE International Conference on Systems,Man,and Cybernetics,Computational Cybernetics and Simulation[C],Orlando,Florida,USA,1997,1:150 ~ 155.
[5]  Shoji Tanaka,Masayuki Inoue,Michitoshi Ishiwaka,et al.A method for extracting and analyzing "Kansei" factors from pictures[A].In:IEEE First Workshop on Multimedia Signal Processing[C],Princeton,New Jersey,USA,1997:251 ~ 256.
[6]  Se Yoon Jeong,Kyuheon Kim,Byung Tae Chun,et al.Entropy and color correlation features for image indexing[A].In:IEEE SMC \' 99 Conference Proceedings[C],Tokyo,Japan,1999,2:895 ~899.
[7]  Nadia bianchi,Luc Berthouze,Toshikazu Kato.Supervised self-organization of user\'s kansei model for image retrieving[A].In:Second International Conference on Cognitive Technology.Humanizing the Information Age Proceedings[C],Aizu-Wakamatsu City,Japan,1997:185 ~ 189.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133