Combining multi-visual features for efficient indexing in a large image database

The optimized distance-based access methods currently available for multidi mensional indexing in multimedia databases have been developed based on two major assumptions: a suitable distance function is known a priori and the dimensionality of the image features is low. It is not trivial to define a distance function that best mimics human visual perception regarding image similarity measurements. Reducing high-dimensional features in images using the popular principle component analysis (PCA) might not always be possibl e due to the non-linear correlations that may be present in the feature vec tors. We propose in this paper a fast and robust hybrid method for non-line ar dimensions reduction of composite image features for indexing in large i mage database. This method incorporates both the PCA and non-linear neural network techniques to reduce the dimensions of feature vectors so that an o ptimized access method can be applied. To incorporate human visual percepti on into our system, we also conducted experiments that involved a number of subjects classifying images into different classes for neural network trai ning. We demonstrate that not: only can our neural network system reduce th e dimensions of the feature vectors, but that the reduced dimensional featu re vectors can also be mapped to an optimized access method for fast and ac curate indexing.