Retrieval and classification of shape-based objects using Fourier, generic Fourier, and wavelet-Fourier descriptors technique: A comparative study

In this paper, we report retrieval and classification of shape-based objects employing three techniques-conventional Fourier descriptors (FD), generic Fourier descriptors (GFD) and wavelet-Fourier descriptors (WFD) techniques. All the three techniques have been applied to a database of seven different types of shapes. The centroid distance based shape signatures have been used for the derivation of descriptors. The Euclidean distance has been calculated as a similarity measure parameter for shape classification. For WFD technique, a Mexican-hat wavelet function was used. Classification results from all the three techniques were compared and it was observed that WFD performs better than FD and GFD technique. To study the effect of the noise on the retrieval and classification of shapes of different objects, additive and multiplicative noise of various variances were applied to the database. Precision and recall were also measured as parameters of performance metric.