The Maximal Overlap Discrete Wavelet Scattering Transform and Its Application in Classification Tasks
This work addresses classification challenges in signal processing domains like ECG analysis, but it is incremental as it combines existing transforms.
The paper tackled the problem of classification tasks with limited training data by introducing the Maximal Overlap Discrete Wavelet Scattering Transform (MODWST), achieving good performance in stationary and ECG signal classification as a viable alternative to CNNs.
We present the Maximal Overlap Discrete Wavelet Scattering Transform (MODWST), whose construction is inspired by the combination of the Maximal Overlap Discrete Wavelet Transform (MODWT) and the Scattering Wavelet Transform (WST). We also discuss the use of MODWST in classification tasks, evaluating its performance in two applications: stationary signal classification and ECG signal classification. The results demonstrate that MODWST achieved good performance in both applications, positioning itself as a viable alternative to popular methods like Convolutional Neural Networks (CNNs), particularly when the training data set is limited.