Page Summary

outlined_flag

• Exponentially Weighted Moving Average Change Detection computes a harmonic model for training data, subtracts it, and analyzes residuals with Shewhart X-bar charts and EWMA to indicate disturbed pixels when exceeding control limits.

• The output is a 5-band image containing EWMA score, harmonic coefficients, RMSE, r-squared, and residuals.

• Negative EWMA values indicate disturbance, and positive values indicate recovery.

Exponentially Weighted Moving Average Change Detection. This algorithm computes a harmonic model for the 'training' portion of the input data and subtracts that from the original results. The residuals are then subjected to Shewhart X-bar charts and an exponentially weighted moving average. Disturbed pixels are indicated when the charts signal a deviation from the given control limits.

 The output is a 5 band image containing the bands:

    ewma: a 1D array of the EWMA score for each input image. Negative values represent disturbance and positive values represent recovery.

    harmonicCoefficients: A 1-D array of the computed harmonic coefficient pairs. The coefficients are ordered as [constant, sin0, cos0, sin1, cos1...]

    rmse: the RMSE from the harmonic regression.

    rSquared: r-squared value from the harmonic regression.

    residuals: 1D array of residuals from the harmonic regression.

See: Brooks, E.B., Wynne, R.H., Thomas, V.A., Blinn, C.E. and Coulston, J.W., 2014. On-the-fly massively multitemporal change detection using statistical quality control charts and Landsat data. IEEE Transactions on Geoscience and Remote Sensing, 52(6), pp.3316-3332.

Usage	Returns
ee.Algorithms.TemporalSegmentation.Ewmacd(timeSeries, vegetationThreshold, trainingStartYear, trainingEndYear,harmonicCount,xBarLimit1,xBarLimit2,lambda,lambdasigs,rounding,persistence)	Image

Argument	Type	Details
timeSeries	ImageCollection	Collection from which to extract EWMA. This collection is expected to contain 1 image for each year and be sorted temporally.
vegetationThreshold	Float	Threshold for vegetation. Values below this are considered non-vegetation.
trainingStartYear	Integer	Start year of training period, inclusive.
trainingEndYear	Integer	End year of training period, exclusive.
harmonicCount	Integer, default: 2	Number of harmonic function pairs (sine and cosine) used.
xBarLimit1	Float, default: 1.5	Threshold for initial training xBar limit.
xBarLimit2	Integer, default: 20	Threshold for running xBar limit.
lambda	Float, default: 0.3	The 'lambda' tuning parameter weighting new years vs the running average.
lambdasigs	Float, default: 3	EWMA control bounds, in units of standard deviations.
rounding	Boolean, default: true	Should rounding be performed for EWMA.
persistence	Integer, default: 3	Minimum number of observations needed to consider a change.