Find Peaks

Finds local maxima in sequential data.

Usage

peaks_find(x, y, ...)

# S4 method for numeric,numeric
peaks_find(x, y, method = "MAD", SNR = 2, m = NULL, ...)

# S4 method for ANY,missing
peaks_find(x, method = "MAD", SNR = 2, m = NULL, ...)

Arguments

x, y: A numeric vector. If y is missing, an attempt is made to interpret x in a suitable way (see grDevices::xy.coords()).
...: Extra parameters to be passed to internal methods.
method: A character string specifying the method to be used for background noise estimation (see below).
SNR: An integer giving the signal-to-noise-ratio for peak detection (see below).
m: An odd integer giving the window size (i.e. the number of adjacent points to be used). If NULL, 5% of the data points is used as the half window size.

Value

Returns a list with two components x and y.

Details

A local maximum has to be the highest one in the given window and has to be higher than \(SNR \times noise\) to be recognized as peak.

The following methods are available for noise estimation:

MAD: Median Absolute Deviation.

Note that to improve peak detection, it may be helpful to smooth the data and remove the baseline beforehand.

Note

There will be \((m - 1) / 2\) points both at the beginning and at the end of the data series for which a complete \(m\)-width window cannot be obtained. To prevent data loss, progressively wider/narrower windows are used at both ends of the data series.

Adapted from Stasia Grinberg's findPeaks function.

Author

N. Frerebeau

Examples

## X-ray diffraction
data("XRD")

## 4S Peak Filling baseline
baseline <- baseline_peakfilling(XRD, n = 10, m = 5, by = 10, sparse = TRUE)

plot(XRD, type = "l", xlab = expression(2*theta), ylab = "Count")
lines(baseline, type = "l", col = "red")


## Correct baseline
XRD <- signal_drift(XRD, lag = baseline, subtract = TRUE)

## Find peaks
peaks <- peaks_find(XRD, SNR = 3, m = 11)

plot(XRD, type = "l", xlab = expression(2*theta), ylab = "Count")
lines(peaks, type = "p", pch = 16, col = "red")
abline(h = attr(peaks, "noise"), lty = 2) # noise threshold


## Half-Width at Half-Maximum
x <- seq(-4, 4, length = 1000)
y <- dnorm(x)

peaks_fwhm(x, y, center = 0) # Expected: 2 * sqrt(2 * log(2))
#> [1] 2.354354