Main Content

AI for Signals and Images

Wavelet-based techniques for machine learning and deep learning, GPU acceleration, hardware deployment, signal labeling

Wavelet techniques are effective for obtaining sparse, compressive data representations or features, which you can use in machine learning and deep learning workflows. Wavelet Toolbox™ supports deployment of multiscale feature extraction algorithms through MATLAB® Coder™ and GPU Coder™ for a number of targets. To take advantage of the performance benefits offered by a modern graphics processing unit (GPU), certain Wavelet Toolbox functions can perform operations on a GPU. These functions provide GPU acceleration for your workflows. Wavelet Toolbox also provides functionality to perform signal labeling.

Highlighted Topics

Categories

  • Working with Signals
    Multiresolution analysis, joint time-frequency scattering, wavelet time scattering, continuous wavelet transform, nondecimated discrete wavelet transform, Wigner-Ville distribution, mel spectrogram
  • Working with Images
    Wavelet image scattering, 2-D continuous wavelet transform, shearlets, stationary wavelet transform
  • GPU Acceleration
    Feature extraction on GPUs for machine learning and deep learning workflows
  • Hardware Deployment
    C/C++ code generation, GPU code generation, Raspberry Pi®, NVIDIA® Jetson®

Related Information

Featured Examples