Searching similar images - Vector quantization with S-tree
نویسندگان
چکیده
Searching of similar pictures was in the past based mainly on searching of similar picture names. We try to find an effective method how to search pictures by searching of similar information in the picture (histograms, shapes, blocks,). There already are some methods but still not effective enough. In this paper we describe a method where we combine vector quantization (VQ) and fuzzy S-trees. Work contains testing of our approach and you can see results in a final chapter of this paper. The benefit of this work is not the final solution but we put a key-stone for further research and for optimizations. First tests show up the efficiency and usefulness of our approach, which is under laid by executed tests. Keywords-vector quantization, image similarity, s-tree, fuzzy sets
منابع مشابه
Competitive Neural Trees for Vector Quantization
This paper presents a self-organizing neural architecture for vector quantization, called Competitive Neural Tree (CNeT). At the node level, the CNeT employs unsupervised competitive learning. The CNeT performs hierarchical clustering of the feature vectors presented to it as examples, while its growth is controlled by a splitting criterion. Because of the tree structure, the prototype in the C...
متن کاملImage Vector Quantization with Hierarchical Tree Structure
The tree-structured vector quantization (TSVQ) is an efficient method for the vector quantization (VQ). In this paper, we propose a new tree structure model, hierarchical tree structure (HTS), to reduce encoding time. Briefly, the HTSVQ decreases the dimension of the vector on the upper levels of the codebook tree. If the vector dimension decreases, then the searching time on the codebook tree ...
متن کاملFractal Compression Using the Discrete Karhunen-Loeve Transform
Fractal coding of images is a quite recent and eecient method whose major drawback is the very slow compression phase, due to a time-consuming similarity search between image blocks. A general acceleration method based on feature vectors is described, of which we can nd many instances in the litterature. This general method is then optimized using the well-known Karhunen-Loeve expansion, allowi...
متن کاملWavelet based Enhanced Color Image Compression relying on Sub-band Vector Quantization
Increase in the use of color images in the continuous expansion of multimedia applications has increased the demand for efficient techniques that can store and transmit visual information. This demand has made image compression a vital factor and has increased the need for efficient algorithms that can result in high compression ratio with minimum loss. This paper proposes an innovative techniq...
متن کاملFast Encoding Algorithm for Vector Quantization
In this paper, we present a new and fast encoding algorithm (FEA) for vector quantization. The magnitude (sum of the components of a vector) feature of the vectors is used in this algorithm to improve the efficiency of searching. Sorting of the magnitude values enhances the searching. As the values are sorted, the searching can be terminated in advance to reduce the time needed to locate the re...
متن کامل