PROJECT TITLE :

Connected Components Objects Feature For Cbir - 2017

ABSTRACT:

With the popularity of the network and enlargement of multimedia technology, the ancient techniques of information retrieval don't satisfy the wants of users. Recently, the content primarily based image retrieval and its techniques have become the new topic to satisfy a nice development. In this paper, a brand new method is proposed to unravel the matter of regions of interest (ROI) primarily based image retrieval. The ROI technique that is predicated on segmenting the image into mounted partitions is computationally costly. The proposed technique relies on the connected parts and attention-grabbing of objects to get the histogram and statistical texture feature vectors. These resulted vectors are used to retrieve pictures from a massive image database. The colour and texture features of the connected elements are computed from the histograms of the quantized HSV color area and Gray Level Co-incidence Matrix (GLCM), respectively. The vectors matching method relies on the histogram intersection. It's obvious the experimental data clearly shows the efficiency of the proposed methodology as compared to the traditional ROI technique in terms of computationally price.