Title :
A classification method of images based on composition and its application to image retrieval
Author :
Maeda, Shigenori ; Mukunoki, Masayuki ; Ikeda, Katsuo
Author_Institution :
Sch. of Inf., Kyoto Univ., Japan
Abstract :
We propose a visual query method for image retrieval, in which the user expresses the composition of the target image by selecting one of the composition types presented by the system, and also propose an image classification method to derive composition types from an image database. From the viewpoint of communication between the system and its user, we point out the problem of visual query methods using concrete images as queries. Though suited to convey visual image properties, they increase the ambiguity in the system´s interpretation of queries. To manage this tradeoff, we implemented methods in a prototype system and derived composition types from an image database to show how our query method works in image retrieval
Keywords :
image classification; image retrieval; visual databases; image classification; image composition; image database; image retrieval; prototype system; query processing; visual image properties; visual query; Clocks; Concrete; Content based retrieval; Image communication; Image databases; Image resolution; Image retrieval; Informatics; Information retrieval; Prototypes;
Conference_Titel :
Multimedia Computing and Systems, 1999. IEEE International Conference on
Conference_Location :
Florence
Print_ISBN :
0-7695-0253-9
DOI :
10.1109/MMCS.1999.778300