Convolution (sofia_redux.toolkit.convolve)

Introduction

The sofia_redux.toolkit.convolve module contains classes and functions to enable N-dimensional convolution with various pre or user-defined kernels. The majority of work is handled by the ConvolveBase class, a child of sofia_redux.toolkit.utilities.base.Model, which is also the parent of the sofia_redux.toolkit.fitting.polynomial.Polyfit class. As such, many features are shared across both classes.

Kernel Convolution

• Convolution
  • Basic Functionality
  • Normalization
  • Masked Values, NaN handling and Error Propagation
  • Robust Outlier Rejection and Statistics for the Convolve Class

Convolution Filters

• Convolution Filters
  • Edge Effects
  • Sobel-Feldman Filtering
  • Savitzky-Golay Filter