Read a capture history csv file (e.g. created in Matlab)

Required columns for each observer are detect_, snr_, signalRMSdB_, noiseRMSdB_, and the underscore is followed by input parameter observerNames. -detect_ columns are binary i.e. 1 a detection and 0 for absence -snr_, signalRMSdB_, and noiseRMSdB_ are all numeric and in dB -groundTruth is the binary column to use as ground-truth detections. It can be a detect_observerName or another column. OG capture history tables have exactly two observers with suffixes, observerNames=c('table1','table2'). These are the type of tables used in the 'Beyond Counting Calls' manuscript. CR capture history tables can have N observers and can have any number of suffixes of any form. MultiCaptureHistoryTable from the common ground manuscript produces suffixes of the form: ('observer1','observerN-1','observerN')

Usage

readCapHist(
  capHistFile,
  observerNames = c("table1", "table2"),
  groundTruth = "table1",
  whichObserver = "detect_table2"
)

Arguments

capHistFile

name of the csv or txt file that contains capture histories. TODO: Document the required columns/file format better

observerNames

list of strings that correspond to the suffixes used for each observer. Default: c('table1','table2')

groundTruth

Name of column with binary data used as ground-truth.

whichObserver

Name of observer for which detection function will be modelled, and call density estimated

Value

capture history table as data.frame