monitoR (version 1.2)

timeAlign: Condense Detections or Peaks from Multiple Templates

Description

Condense detections or peaks from a number of templates (of the same detection type); events that occur within an adjustable time buffer of one another are assumed to be duplicate detections. In such cases the event with the highest score is saved. Functions with detections for a single species or multiple species.

Usage

timeAlign(x, what = "detections", tol = 1)

Value

Returns a single data frame of detections (the input x) with duplicated events removed, leaving only the event that had the highest score.

Arguments

x

An object of class detectionList, a single data frame of detections, or list of either file paths to a csv file or of data frames.

what

Character, in c("detections", "peaks"). Detections are peaks above a score cutoff. Peaks are all peaks. Required only if x is of class detectionList

tol

Numeric value for tolerance, with units seconds. If a detected event is within this value (actually +/- 0.5tol), the events are assumed to co-occur and be of the same origin. A somewhat arbitrary value (like epsilon), but should be less than 2/3 the template duration.

Author

Jon Katz

Details

If input is an object of class detectionList, a single data frame, or list of either file paths or data frames. Must be called for each survey.

See Also

The function eventEval operates similarly, but rather than merge detection results from multiple templates it compares them to known events and reports the True +, True -, False +, and False - rates.

Examples

if (FALSE) {
# Not run because it will create files in user's working directory
data(survey)
data(btnw)

writeWave(btnw, "btnw.wav")

btnw2 <- cutw(survey, from = 0.75, to = 3)

writeWave(btnw2, "btnw2.wav")

# Template construction
btnw1 <- makeBinTemplate(
        "btnw.wav", 
        frq.lim = c(2, 8), 
        select = "auto", 
        name = "btnw1", 
        buffer = 4, 
        amp.cutoff = -31, 
        binary = TRUE)

btnw2 <- makeBinTemplate(
        "btnw2.wav", 
        frq.lim = c(2, 8), 
        select = "auto", 
        name = "btnw2", 
        buffer = 4, 
        amp.cutoff = -24, 
        binary = TRUE)

# Join templates
btnw <- combineBinTemplates(btnw1, btnw2)

# Binary point matching
scores <- binMatch(survey = survey, templates = btnw, time.source = 'fileinfo')

# Isolate peaks
pks <- findPeaks(scores)

# View detections
getDetections(pks)

# Compare to output of timeAlign
timeAlign(pks)}