Low-Level Features for Image Retrieval Based on Extraction of Directional Binary Patterns and Its Oriented Gradients Histogram

摘要

In this paper, we present a novel approach for image retrieval based onextraction of low level features using techniques such as Directional BinaryCode, Haar Wavelet transform and Histogram of Oriented Gradients. The DBCtexture descriptor captures the spatial relationship between any pair ofneighbourhood pixels in a local region along a given direction, while LocalBinary Patterns descriptor considers the relationship between a given pixel andits surrounding neighbours. Therefore, DBC captures more spatial informationthan LBP and its variants, also it can extract more edge information than LBP.Hence, we employ DBC technique in order to extract grey level texture featurefrom each RGB channels individually and computed texture maps are furthercombined which represents colour texture features of an image. Then, wedecomposed the extracted colour texture map and original image using Haarwavelet transform. Finally, we encode the shape and local features of wavelettransformed images using Histogram of Oriented Gradients for content basedimage retrieval. The performance of proposed method is compared with existingmethods on two databases such as Wang's corel image and Caltech 256. Theevaluation results show that our approach outperforms the existing methods forimage retrieval.