ABSTRACT

This paper presents the method developed to search and retrieve similar images from a large database. The Fourier transform is used to generate the feature vectors based on the mean values of real and imaginary parts of complex numbers of polar coordinates in frequency domain. 8 mean values of 4 upper half sectors real and imaginary parts of each R, G and B components of an image are considered for feature vector generation. The algorithm uses 24 mean values of real and imaginary parts in total. Euclidian distances between the feature vectors of query image and the database images are considered. Images are retrieved in ascending order of Euclidian distances. The Average precision and Average recall of each class and overall average of all averages of each class are calculated as a performance measure.