Local Oppugnant Color Texture Pattern for image retrieval system

The current scenario of image retrieval pays attention to local texture patterns. The recently proposed Local Tetra Pattern (LTrP) represents the image by the directional information and gives promising results. This paper proposes Local Oppugnant Color Texture Pattern (LOCTP), an enhancement of LTrP, which is able to discriminate the information derived from spatial inter-chromatic texture patterns of different spectral channels within a region. It determines the relationship in terms of the intensity and directional information between the referenced pixels and their oppugnant neighbors. The LOCTP strives to use the harmonized link between color and texture, which helps the system to incorporate the human perception. The experimental analysis of the proposed method is done with state-of-art techniques by using standard image databases Brodatz texture database (DB1) and Corel database (DB2). Also, the evaluation has been done in various color models like YCbCr, HSV, Lab, and RGB. In addition, a feature-level fusion framework is used to combine the Colored Pattern Appearance Model (CPAM) and the LOCTP for getting better result in natural images. The experimental results show considerable improvement in terms of average precision, average recall and average retrieval rate when compared with the previous works. (C) 2014 Elsevier B.V. All rights reserved.