TY - GEN
T1 - Robust Image Retrieval Based on Mixture Modeling of Weighted Spatio-color Information
AU - Luszczkiewicz-Piatek, Maria
AU - Smolka, Bogdan
PY - 2015
Y1 - 2015
N2 - In this paper we propose a novel approach to color image retrieval. Color information is modeled using Gaussian mixtures and incorporates the information on the spatial distribution of the color image pixels utilizing the Dijkstra algorithm. The proposed algorithm has high indexing performance and operates on model of low dimensionality. Thus, the proposed method needs only the adjustment of Gaussian Mixture Model parameters for efficient color image retrieval. The proposed method is extensively tested on Corel and Wang dataset. The results demonstrate that proposed framework is more efficient than other methods when images are subjected to lossy coding such as JPEG method.
AB - In this paper we propose a novel approach to color image retrieval. Color information is modeled using Gaussian mixtures and incorporates the information on the spatial distribution of the color image pixels utilizing the Dijkstra algorithm. The proposed algorithm has high indexing performance and operates on model of low dimensionality. Thus, the proposed method needs only the adjustment of Gaussian Mixture Model parameters for efficient color image retrieval. The proposed method is extensively tested on Corel and Wang dataset. The results demonstrate that proposed framework is more efficient than other methods when images are subjected to lossy coding such as JPEG method.
UR - https://www.scopus.com/pages/publications/84907378687
U2 - 10.1007/978-3-319-10662-5_11
DO - 10.1007/978-3-319-10662-5_11
M3 - Conference contribution
AN - SCOPUS:84907378687
SN - 9783319106618
T3 - Advances in Intelligent Systems and Computing
SP - 85
EP - 93
BT - Image Processing and Communications Challenges 6
PB - Springer Verlag
T2 - 6th International Image Processing and Communications Conference, IPandC 2014
Y2 - 10 September 2014 through 12 September 2014
ER -