Discrete Fourier transform based pattern classifiers

Abstract

Abstract.A technique for pattern classification using the Fourier transform combined with the nearest neighbor classifier is proposed. The multidimensional fast Fourier transform (FFT) is applied to the patterns in the data base. Then the magnitudes of the Fourier coefficients are sorted in descending order and the first P coefficients with largest magnitudes are selected, where P is a design parameter. These coefficients are then used in further processing rather than the original patterns. When a noisy pattern is presented for classification, the pattern's P Fourier coefficients with largest magnitude are extracted. The coefficients are arranged in a vector in the descending order of their magnitudes. The obtained vector is referred to as the signature vector of the corresponding pattern. Then the distance between the signature vector of the pattern to be classified and the signature vectors of the patterns in the data base are computed and the pattern to be classified is matched with a pattern in the data base whose signature vector is closest to the signature vector of the pattern being classified.

Robust Object Classification Approach Using Spherical Harmonics

“…Initial spherical harmonics [13], [14] or Fourier [57], [58] based classifiers used the magnitudes of the coefficients to identify similarities between images. Those magnitudes were used because they are rotation invariant [13], have low dimensions and importantly they are useful for building robust classification techniques [57], [58]. The noise mitigation property was achieved by discarding the coefficients that are greatly affected by noise.…”

“…The noise mitigation property was achieved by discarding the coefficients that are greatly affected by noise. For instance, in [58], only frequencies with high magnitudes were used, while in [57], only low-frequency components were used.…”

Discretization of data using Boolean transformations and information theory based evaluation criteria

“…However, the primary goal is not to acquire the data itself, but to extract the knowledge from it, e.g., patterns or rules that would allow better systematization and explanation of the observed phenomena. The answer to these needs is knowledge discovery in databases (KDD), described in more detail in [1], as well as many examples of knowledge retrieval methods that applied in processed data can help in decision making [2][3][4][5].…”

Material recognition by feature classification using time-of-flight camera

“…In general, the FFT is a powerful tool for pattern recognition. It is commonly employed to extract invariant features 24 because of its important properties; for example, a shift in the time domain does not involve any change in the amplitude spectrum of the image. Good predictive accuracies in material recognition are expected since the frequency domain representation might provide more useful information than the time domain one.…”