preprocess: Preprocess Gaze Data

In maltelueken/gazeHMM: Classify Gaze Data With a Hidden Markov Model

Description

Prepares raw gaze data for classification by a hidden Markov model as part of gazeHMM.

Usage

preprocess(
  x,
  y,
  t,
  unit = "px",
  res,
  dim,
  dist,
  fr,
  blink = NULL,
  b.win = 0.05,
  sg.order = 3,
  sg.length = sg.order + 3 - sg.order%%2
)

Arguments

x	Numeric vector of raw x coordinates for each gaze samples.
y	Numeric vector of raw y coordinates for each gaze samples.
t	Numeric vector of time stamps for each gaze sample.
unit	Character string indicating the unit of coordinates (either 'px' or 'va').
res	Screen resolution (in px).
dim	Screen dimensions (in mm).
dist	Distance between subject and screen (in mm).
fr	Sampling rate of the eye-tracker (in Hz).
blink	Either a numeric vector of length two indicating the x and y coordinates for blink samples or a logical vector with the same length as x, y, and t indicating blink samples.
b.win	Time window around blink samples that are to be ignored (i.e., set to NA; in s).
sg.order	Order of the Savitzky-Golay filter.
sg.length	Length of the Savitzky-Golay filter (must be odd).

Value

A data frame with x and y gaze coordinates in degrees of visual angle, time stamps, velocity, accelelration, sample-to-sample angle, and an initial label for each sample.

maltelueken/gazeHMM documentation built on Oct. 6, 2020, 11:10 a.m.