Semantic modeling of natural scenes by local binary pattern

Abstract

Automatic image annotation is an efficient and promising solution in content based image retrieval system applications to process very large databases via keywords. The basic idea of semantic modeling is to describe local image regions into semantic concepts using low level features such as color and texture. These local image region descriptions are combined to a global image representation that can be used for scene categorization and retrieval. In this paper, Local Binary Pattern features and neighborhood prior information are used as texture and spatial features for local image representation that allows access to natural scenes. K-Means classifier has been used to support automatic image annotation of local image region into semantic classes such as water, sky, and trees. Extensive experiments on databases like COREL, shows that the proposed technique performs well in scene classification.