Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Clustering and dimensionality reduction for image retrieval in high-dimensional spaces

Benkrama S., Zaoui L., Charrier C.

Journal of Theoretical and Applied Computer Science

2014

Vol. 8, nr 3

37--50

The scalability of indexing techniques and image retrieval pose many problems. Indeed, their performance degrades rapidly when the database size increases. In this paper, we propose an efficient indexing method for high-dimensional spaces. We investigate how high-dimensional indexing methods can be used on a partitioned space into clusters to help the design of an efficient and robust CBIR scheme. We develop a new method for efficient clustering is used for structuring objects in the feature space; this method allows dividing the base into data groups according to their similarity, in function of the parameter threshold and vocabulary size. A comparative study is presented between the proposed method and a set of classification methods. The experiments results on the Pascal Visual Object Classes challenges (VOC) of 2007 and Caltech-256 dataset show that our method significantly improves the performance. Experimental retrieval results based on the precision/recall measures show interesting results.

A classified vector quantization scheme for color image coding perceptually tuned

Charrier C., Cherifi H.

Image Processing & Communications

2001

Vol. 7, no 1-2

17-29

Edges are of fundamental importance in the analysis of images, and of course in the field of image quality. To incorporate the edgy: information as coded by the Human Visual System (HVS) in a vector quantization scheme, we have developed a classification strategy to separate edge vectors from non-edge vectors. This strategy allows the generation of different sets of codewords different size for each kind of vectors. For each one of the "edge" sets, the final size is perceptually tuned. Finally, when an image is encoded, its associated edge map is generated. Then the selection of the appropriate "edge" set is made in respect with the edge amount present in the image. Then the second set of non-edge vectors is performed in order to respect the required compression rate. Statistical measure and psychophysical experiments have been performed to judge the quality of reconstructed images.