Image Retrieval System Based on Feature Computation – an Integrated Approach

The idea proposed in this paper utilizes an efficient image retrieval system that is based on computation and combination of grey level, shape and texture features. The proposed system calculates various grey level features which are used to extract attributes from an input image. Grey Level Co-occurrence Matrix is used to compute the texture feature of a given image while the Shape feature finds the shape of the specified object. Scale Invariant Feature Transform Algorithm is incorporated to extract the feature points of an image to detect the specific object among various kinds of image classes. The similarity of image objects is assessed by combining these features that are computed already. The “sample” database is constructed with different kinds of image classes such as car, bike etc. Keywords-Scale Invariant Feature Transform, Grey Level Co-occurrence Matrix, Image Retrieval.