figure_barPlot: Detect and display all bar plot objects.

In metagear: Comprehensive Research Synthesis Tools for Systematic Reviews and Meta-Analysis

Description

Automated detection of grouped data displayed in a bar-plot/chart figure image. The default returns these detected objects as an EBImage raster image, and as a vector of all the estimated lengths that are proportional to the values presented on each bar (and their error bars, if they are present). Note that the extracted points will be sorted by their positioning on the X-axis (or Y if the plot is a horizontal bar plot). For example, if there were error bars in the figure these will be grouped with the detected bar column. However, within these X-axis positioning they will not be sorted. See vignette for worked several illustrations.

Usage

figure_barPlot(
  file = file.choose(),
  horizontal = FALSE,
  binary_threshold = 0.6,
  axis_thickness = 3,
  axis_sensitivity = 0.2,
  axis_length = 0.75,
  axis_X_color = "#00ABAB",
  axis_Y_color = "#B0D36A",
  Y_min = 0,
  Y_max = 100,
  bar_width = 9,
  bar_sensitivity = 0.1,
  point_color = "#0098B2",
  point_size = 9,
  ignore = FALSE
)

Arguments

file	The file name and location of a bar-plot figure. Prompts for file name if none is explicitly called.
horizontal	If TRUE then aims to detect objects from a bar-plot that depicts data horizontally (rather than vertically).
binary_threshold	A proportion from zero to one designating the gray-scale threshold to convert pixels into black or white. Pixel intensities below the proportion will be converted to black, and those above white.
axis_thickness	An integer used to designate the thickness of the axis lines on a figure. Close alignment to the thickness of the axis on a figure will improve axis detection.
axis_sensitivity	A value designating the sensitivity of identifying straight lines on figure. A smaller number results in a higher sensitivity to identify axes.
axis_length	The relative size of the axis to the figure. The default is that axis lengths are 0.75 (75 percent) the size of the figure. This option is necessary since bar lengths may be similar to the axis length. Values should range between zero and one.
axis_X_color	The color to paint the detected X-axis.
axis_Y_color	The color to paint the detected Y-axis.
Y_min	The minimum Y value displayed on the Y-axis (used to scale detected data points).
Y_max	The maximum Y value displayed on the Y-axis (used to scale detected data points).
bar_width	An integer value designating the width of vertical lines on bars. A smaller number should be used when the width of bars are small (as well as the width of error bars).
bar_sensitivity	A value designating the sensitivity of identifying the vertical lines on bars. A smaller number should be used when the thickness of bars are small (as well as the width of error bars).
point_color	The color to paint the circles identifying the detected levels on bar columns and error bars.
point_size	An integer used to designate the size of the points painting the detected bars on a figure.
ignore	When TRUE does not display painted image with detections, only returns the data frame with detected points.

Value

A vector of scaled lengths for detected column and error bars.

metagear documentation built on Feb. 15, 2021, 5:09 p.m.