Title :
Combined feature descriptor for Content based Image Retrieval
Author :
Selvarajah, S. ; Kodithuwakku, S.R.
Author_Institution :
Dept. of Phys. Sci., Univ. of Jaffna, Jaffna, Sri Lanka
Abstract :
Content based Image Retrieval (CBIR) allows automatically extracting target images according to objective visual contents of the image itself. Representation of visual features and similarity match are important issues in CBIR. Colour and texture features are important properties in CBIR systems. In this paper, a combined feature descriptor for CBIR is proposed to enhance the retrieval performance for CBIR. This method is developed by exploiting the wavelets and colour histogram moments. First, Haar wavelet is used to decompose colour images into wavelet coefficients. Then image feature extraction and similarity matching are performed by means of histogram moments. Ten categories of colour images are used to test the proposed technique. Experiment results show significant improvement of retrieval efficiency as compared to that of histogram moments as well as 2D Discrete Wavelet Transform.
Keywords :
Haar transforms; content-based retrieval; discrete wavelet transforms; feature extraction; image colour analysis; image retrieval; CBIR; Haar wavelet; colour histogram; content based image retrieval; discrete wavelet transform; feature descriptor; histogram moments; image color decomposition; image feature extraction; target image extraction; visual contents; visual feature representation; wavelet coefficients; Discrete wavelet transforms; Feature extraction; Histograms; Image color analysis; Image retrieval; 2D Discrete Wavelet Transform; Content based Image Retrieval; Haar wavelet; colour histogram moments;
Conference_Titel :
Industrial and Information Systems (ICIIS), 2011 6th IEEE International Conference on
Conference_Location :
Kandy
Print_ISBN :
978-1-4577-0032-3
DOI :
10.1109/ICIINFS.2011.6038060
