Image retrieval via geometrical shape moments

Abstract EN

Object shape representation plays an important role in the area of image processing, pattern recognition and computer vision.

A set of moments can form a global shape descriptor of an image.

Different spatial characteristics can be represented by different geometric moments (GM).

This paper deals with the content based image retrieval (CBIR) using geometric moments for features extraction.

The GM moments are the simplest among moment functions, of the kernel function defined as a product of the pixel coordinates.

The main advantage with the GMs is that image coordinate transformations can be easily expressed and analyzed in terms of the corresponding transformations in the moment space.

The proposed method has a fast training and test times compared with color coherent and texture image retrieval.

The proposed method utilizes the active learning in the training step.