Fuzzy Aggregation Of Image Features In Content-Based Image Retrieval
ABSTRACT The obstacle of generating hybrid queries within the context of content-based image retrieval is still very real. In attempts to overcome this, fuzzy aggregation can be used to combine single, simple index queries into larger, more complex ones. This paper outlines the use of a fuzzy aggregation technique for hybrid querying which has the ability to adjust its behavior according operator-controlled parameters. The resulting aggregator can be viewed as a featureadaptive overall similarity measure. For the purposes of this extended summary, the scope of the aggregator is limited to queries involving color content, color coverage, and horizontal /vertical trends, and applied to a media database comprised of COREL images of fixed size. Preliminary results show promise and illustrate that hybrid queries using the aforementioned fuzzy aggregator are effective in their ability to retrieve relevant images while suppressing erroneous retrievals when compared to simple, single-feature queries. In addition, the results obtained are at a minimum compara- ble to multiple-feature queries generated using a weighted mean approach but exhibiting scalability and greater fiexir bility in parameter adjustment.
Conference Paper: OWA fuzzy linking histogram approach for image retrieval[Show abstract] [Hide abstract]
ABSTRACT: Color image histograms are very useful tools for content based image retrieval (CBIR) that can be applied on features such as colour, texture and shape. As these kinds of histograms results with large variations between neighbouring bins, they seem so sensitive to any kind of changes such as noise, illumination. To overcome this problem, in this paper, fuzzy linking histogram approach based on OWA aggregation operator is proposed, which is capable of projecting 3-dimensional (L*a*b*) colour histograms into single-dimension. The proposed method have been evaluated and compared with five other related methods in retrieving similar images from the common dataset which is available on http://utopia.duth.gr/~konkonst. The experimental results on 100 images within two categories of Cat and Sky reveals better performance of the proposed method in comparison with the other mentioned methods.Computational Cybernetics, 2009. ICCC 2009. IEEE International Conference on; 03/2009