Skip to contents

Utilities for object measures

Source: R/utils_measures.R
utils_measures.Rd

  • get_measures() computes object measures (area, perimeter, radius) by using either a known resolution (dpi) or an object with known measurements.

  • plot_measures() draws the object measures given in an object to the current plot. The object identification ("id") is drawn by default.

Usage

get_measures(
  object,
  measure = NULL,
  id = NULL,
  dpi = NULL,
  sep = "\\_|-",
  verbose = TRUE,
  digits = 5
)

plot_measures(
  object,
  measure = "id",
  id = NULL,
  hjust = NULL,
  vjust = NULL,
  digits = 2,
  size = 0.9,
  col = "white",
  ...
)

Arguments

object

An object computed with analyze_objects().

measure

For plot_measures(), a character string; for get_measures(), a two-sided formula, e.g., measure = area ~ 100 indicating the known value of object id. The right-hand side is the known value and the left-hand side can be one of the following.

  • area The known area of the object.

  • perimeter The known perimeter of the object.

  • radius_mean The known radius of the object.

  • radius_min The known minimum radius of the object. If the object is a square, then the radius_min of such object will be L/2 where L is the length of the square side.

  • radius_max The known maximum radius of the object. If the object is a square, then the radius_max of such object according to the Pythagorean theorem will be L x sqrt(2) / 2 where L is the length of the square side.

id

An object in the image to indicate a known value.

dpi

A known resolution of the image in DPI (dots per inch).

sep

Regular expression to manage file names. The function combines in the merge object the object measures (sum of area and mean of all the other measures) of all images that share the same filename prefix, defined as the part of the filename preceding the first hyphen (-) or underscore (_) (no hyphen or underscore is required). For example, the measures of images named L1-1.jpeg, L1-2.jpeg, and L1-3.jpeg would be combined into a single image information (L1). This feature allows the user to treat multiple images as belonging to a single sample, if desired. Defaults to sep = "\\_|-".

verbose

If FALSE, runs the code silently.

digits

The number of significant figures. Defaults to 2.

hjust, vjust

A numeric value to adjust the labels horizontally and vertically. Positive values will move labels to right (hjust) and top (vjust). Negative values will move the labels to left and bottom, respectively.

size

The size of the text. Defaults to 0.9.

col

The color of the text. Defaults to "white".

...

Further arguments passed on to graphics::text().

Value

  • For get_measures(), if measure is informed, the pixel values will be corrected by the value of the known object, given in the unit of the right-hand side of meae. If dpi is informed, then all the measures will be adjusted to the knosurwn dpi.

  • If applied to an object of class anal_obj, returns a data frame with the object id and the (corrected) measures.

    • If applied to an object of class anal_obj_ls, returns a list of class measures_ls, with two objects: (i) results, a data frame containing the identification of each image (img) and object within each image (id); and (ii) summary a data frame containing the values for each image. If more than one object is detected in a given image, the number of objects (n), total area (area_sum), mean area (area_mean) and the standard deviation of the area (area_sd) will be computed. For the other measures (perimeter and radius), the mean values are presented.

  • plot_measures() returns a NULL object, drawing the text according to the x and y coordinates of the objects in object.

Author

Tiago Olivoto tiagoolivoto@gmail.com

Examples

if (interactive() && requireNamespace("EBImage")) {
library(pliman)
img <- image_pliman("objects_300dpi.jpg")
plot(img)
# Image with four objects with a known resolution of 300 dpi
# Higher square: 10 x 10 cm
# Lower square: 5 x 5 cm
# Rectangle: 4 x 2 cm
# Circle: 3 cm in diameter

# Count the objects using the blue band to segment the image
results <-
   analyze_objects(img,
                 index = "B",
                 lower_noise = 0.1)
plot_measures(results, measure = "id")

# Get object measures by declaring the known resolution in dots per inch
(measures <- get_measures(results, dpi = 300))

# Calculated diagonal of the object 1
# 10 * sqrt(2) = 14.14

# Observed diagonal of the object 1
measures[1, "radius_max"] * 2


# Get object measures by declaring the known area of object 1
get_measures(results,
             id = 1,
             area ~ 100)
}


#> -----------------------------------------
#> measures corrected with:
#> object id: 1
#> area     : 100
#> -----------------------------------------
#> Total    : 40.001 
#> Average  : 13.334 
#> -----------------------------------------