monitor_sop

Notedocblock 
monitor_sop(data, lam, cl, d1, d2; chart_choice=TauTilde(), refinement=OrdinaryType(),
  add_noise=false, noise_dist=Uniform(0, 1))

Monitor the image sequence data (a 3D array, rows × columns × time) with the EWMA control chart based on spatial ordinal patterns and return a ControlChartResult with the chart statistics, the control limit and the first alarm.

The statistics are those of stat_sop(data, lam, d1, d2; ...). All four monitoring statistics are two sided, so the chart signals when abs(stat) > cl.

  • data::Array{<:Real,3}: the image sequence.
  • lam::Float64: EWMA smoothing parameter.
  • cl::Float64: control limit, typically obtained from cl_sop.
  • d1::Int, d2::Int: row and column delays.
  • chart_choice: one of TauHat(), KappaHat(), TauTilde(), KappaTilde().
  • refinement: OrdinaryType() for the classical SOP classification, or one of RotationType(), DirectionType(), DiagonalType(). Must be the same as in the cl_sop call that produced cl.
  • add_noise::Bool=false, noise_dist: add noise to break ties, as in stat_sop.
images = randn(11, 11, 40)
res = monitor_sop(images, 0.1, 0.03, 1, 1; chart_choice=TauTilde())