東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Keyblock: An approach for content-ba...

Zhu, Lei.

FindBook

Google Book

Amazon

博客來

Keyblock: An approach for content-based image retrieval.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Keyblock: An approach for content-based image retrieval./
作者:	Zhu, Lei.
面頁冊數:	296 p.
附註:	Source: Dissertation Abstracts International, Volume: 62-08, Section: B, page: 3697.
Contained By:	Dissertation Abstracts International62-08B.
標題:	Computer Science. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3021963
ISBN:	049334389X

Keyblock: An approach for content-based image retrieval.
Zhu, Lei.

Keyblock: An approach for content-based image retrieval. - 296 p.

Source: Dissertation Abstracts International, Volume: 62-08, Section: B, page: 3697.

Thesis (Ph.D.)--State University of New York at Buffalo, 2001.

The success in text-based information retrieval motivates us to investigate analogous techniques which can support querying and browsing of image data. However, images are greatly different from text both syntactically and semantically in representing and expressing information. Thus, the generalization of information retrieval from the text domain to the image domain is non-trivial.

ISBN: 049334389XSubjects--Topical Terms:

626642
Computer Science.

Keyblock: An approach for content-based image retrieval.
LDR:02895nmm 2200313 4500 001 1810510
005 20040311141527.5
008 130610s2001 eng d
020 $a 049334389X
035 $a (UnM)AAI3021963
035 $a AAI3021963
040 $a UnM $c UnM
100 1 $a Zhu, Lei. $3 1288434
245 1 0 $a Keyblock: An approach for content-based image retrieval.
300 $a 296 p.
500 $a Source: Dissertation Abstracts International, Volume: 62-08, Section: B, page: 3697.
500 $a Major Professor: Aidong Zhang.
502 $a Thesis (Ph.D.)--State University of New York at Buffalo, 2001.
520 $a The success in text-based information retrieval motivates us to investigate analogous techniques which can support querying and browsing of image data. However, images are greatly different from text both syntactically and semantically in representing and expressing information. Thus, the generalization of information retrieval from the text domain to the image domain is non-trivial.
520 $a This dissertation presents a framework of information retrieval for the image domain to support content-based querying and browsing on images. To establish such a theory and techniques, a critical problem is how to construct a codebook of “keywords” for images, which is analogous to the dictionary for text documents. We term such “keywords” in the image domain as “keyblocks”.
520 $a To introduce the framework of keyblock approach for content-based image retrieval, we will first present various keyblock generation approaches, then focus on keyblock-based feature extraction and image retrieval.
520 $a The key idea of our approach is: with keyblocks generated by clustering algorithms, each image can be encoded as 1-dimensional index codes of the keyblocks in the codebook, which is similar to a text document that is considered as a linear list of keywords in text-based information retrieval. Since images have similar representation formats as the text documents, we hope the representation models as well as the retrieval techniques used in the text-based information retrieval can still be effective for the image retrieval.
520 $a In this dissertation, we have generalized various image representation models from the text domain and conducted comprehensive experiments based on test-beds we have set up. By comparing the performance of our approach with conventional techniques using color features and wavelet texture features, the experimental results demonstrate that the keyblock approach is novel and effective in the content-based image retrieval.
590 $a School code: 0656.
650 4 $a Computer Science. $3 626642
690 $a 0984
710 2 0 $a State University of New York at Buffalo. $3 1017814
773 0 $t Dissertation Abstracts International $g 62-08B.
790 1 0 $a Zhang, Aidong, $e advisor
790 $a 0656
791 $a Ph.D.
792 $a 2001
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3021963