2D Discrete Fast Fourier Transform with variable parameters

A Mutual-Attention Guided Feature Extraction and Adaptative Decision Fusion Framework for Fine-Grained Dual-Band Radar Target Classification

“…The physical features include the location and magnitude of peak, the scattering center location, the intensity magnitude, the profile, the perimeter and area of the target, the shape, the micro-motion characteristics, etc. The transform domain features mean the features obtained by Fourier Transform [15], [16], Wavelet Transform [17], [18], Non-negative Matrix Factorization [19], [20], Radon Transform [21], or Sparse Representation [22], [23], etc. Then, these features are classified by K-nearest Neighbor Classifier [24], Bayesian Classifier [25], AdaBoosting Classifier [26], [27], Support Vector Machine (SVM) [28], or Hidden Markov Model [29], etc.…”

Atrial Fibrillation Detection with Single-Lead Electrocardiogram Based on Temporal Convolutional Network–ResNet

“…The progress begins with the segmentation of raw ECG signals from the original dataset.Then, the dataset is divided into training set and test set. In addition, spectrograms are calculated from the raw ECG signal, using the fast Fourier transform[19]. As a result, ECG singals and spectrograms are both input into the classification model, which belongs to the core part of the algorithm.…”