Title :
Directional detail histogram for image retrieval
Author :
Androutsos, D. ; Plataniotis, Konstantinos N. ; Venetsanopoulos, A.N.
Author_Institution :
Dept. of Electr. & Comput. Eng., Toronto Univ., Ont.
fDate :
11/6/1997 12:00:00 AM
Abstract :
The authors propose a new indexing technique, for image retrieval, which calculates the histogram of the directional detail content in a given image. A multiresolution analysis is applied to extract directional information which is then mapped into three-dimensional vectors and presented as a histogram. This allows the use of histogram query techniques to be readily applied for retrieval
Keywords :
image matching; image resolution; indexing; visual databases; wavelet transforms; directional detail histogram; histogram query techniques; image retrieval; indexing technique; multiresolution analysis; three-dimensional vectors;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:19971352