Extraction of Pertinent Subsets from Time-Frequency Representations for Detection and Recognition Purposes

Benoît Leprettre (1) , Nadine Martin (1)

1 LIS - Laboratoire des images et des signaux

A time-frequency representation can highlight non-stationarities in a signal. We propose to extract subsets from the Time-Frequency Representation (TFR) for classification or recognition purposes. We developed two approaches. The first one is developed for TFRs obtained from the Short Time Fourier Transform or the gliding Minimum Variance method. The extraction of compact subsets is viewed as a segmentation of the TFR, which is performed by morphological filtering and Watershed segmentation. The second approach is developed when the TFR has been obtained using parametric estimators. We consider a hybrid estimator, the ARCAP method, and use a Kalman filter trajectory tracker to extract spectral lines. The proposed methods are illustrated by examples on natural signals : dolphin whistle acoustical signals, cavitation signals and seismic signals produced by snow avalanches.

Mots clés: Non-stationary signal interpretation; Time-frequency representation; Watershed segmentation; Kalman Filter; Trajectory tracking; Capon Filter; Minimum variance estimator; Autoregressive model; Spectral analysis

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Benoît Leprettre, Nadine Martin. Extraction of Pertinent Subsets from Time-Frequency Representations for Detection and Recognition Purposes. Signal Processing, 2002, pp.229-238. ⟨hal-00944753⟩