ABSTRACT
In content-based image retrieval (CBIR) applications, the idea of indexing is mapping the extracted descriptors from images into a high-dimensional space. This chapter considers visual features like color, texture, and shape. After combining color, texture, and shape features, the feature vector is reduced using kernel principle component analysis (KPCA). Then, the KD-tree is used for indexing the images. A new proposed optimized range search (ORS) algorithm is used to obtain the optimal range for retrieving the relevant images from the database. The proposed KD-ORS tree is compared with the other existing trees. Hill climbing segmentation begins with identifying the salient regions. Saliency is determined as the local contrast of an image region with respect to its neighborhood at various scales. Typically, there are two types of visual features in CBIR: primitive features that include color, shape, and texture, and domain specific features. Texture means visual pattern, which is one of the main features utilized in image processing.
