CBIR Based on Adaptive Segmentation of HSV Color Space

Author

An, Youngeun ; Riaz, Muhammad ; Park, Jongan

Author_Institution

Dept. of Inf. & Commun. Eng., Chosun Univ., Gwangju, South Korea

fYear

2010

fDate

24-26 March 2010

Firstpage

248

Lastpage

251

Abstract

Proposed algorithm is based on color information using HSV color space. Histogram search characterizes an image by its color distribution, or histogram but the drawback of a global histogram representation is that information about object location, shape, and texture is discarded. Thus local histogram is used for extracting the maximum color occurrence from each segment. Before extracting the maximum color from each segment the input image is adaptively segmented. Different quantization of hue and saturation are used for partitioning the image into different number of segments. Finally minkowski metric is used for feature vector comparison. Web based image retrieval demo system is built to make it easy to test the retrieval performance and to expedite further algorithm investigation.

Keywords

content-based retrieval; feature extraction; image colour analysis; image retrieval; image segmentation; image texture; quantisation (signal); HSV color space adaptive segmentation; Web based image retrieval demo system; color distribution; content based image retrieval; global histogram representation; histogram distribution; hue quantization; image partitioning; image segmentation; image texture; local histogram; maximum color occurrence extraction; minkowski metric; saturation quantization; Data mining; Feature extraction; Histograms; Humans; Image databases; Image processing; Image retrieval; Image segmentation; Information retrieval; Spatial databases; Adaptive Segmentation; Feature Extraction; HSV Color Space; Image Retrieval; Maximum Color;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Modelling and Simulation (UKSim), 2010 12th International Conference on

Conference_Location

Cambridge

Print_ISBN

978-1-4244-6614-6

Type

conf

DOI

10.1109/UKSIM.2010.53

Filename

5481172