Skip to contents

Coerce to Momocs morphometric classes

Source: R/classes_and_printers.R
as_class.Rd

Add class attributes to objects for proper S3 dispatch and printing in morphometric workflows.

Usage

as_coo(x)

as_out(x)

as_ldk(x)

as_xy(x)

as_ldk_id(x)

as_cur(x)

as_path(x)

as_meas(x)

as_coe(x)

as_eft(x)

as_rft(x)

as_dct(x)

as_npoly(x)

as_opoly(x)

as_proc(x)

Arguments

x

An object to coerce (typically a matrix, vector, or list).

Value

Object with updated class attribute.

Details

These functions add Momocs-specific class attributes that enable:

  • Custom printing methods

  • Pretty display in tibbles (via pillar)

  • S3 method dispatch for morphometric operations

  • Type checking and validation

Coordinate classes (inherit from "coo")

Used for shape coordinate data, typically stored as nx2 matrices or list-columns of matrices:

  • as_coo(): Generic coordinate object - base class for all coordinate types

  • as_out(): Closed outlines - class c("out", "coo"). For shapes where the first and last points connect (e.g., leaf outlines, bottle silhouettes)

  • as_ldk(): Landmarks - class c("ldk", "coo"). For discrete anatomical points (e.g., skull landmarks, wing vein intersections)

  • as_cur(): Open curves - class c("cur", "coo"). For shapes with distinct start and end points (e.g., leaf midribs, antenna segments)

Single coordinate matrix

  • as_xy(): Single coordinate matrix - class "xy". Used for individual shape matrices (not list-columns). Has custom print method showing dimensions

Landmark identifiers

  • as_ldk_id(): Landmark indices - class "ldk_id". Integer vectors indicating which points are landmarks. Does NOT inherit from "coo" since it contains indices, not coordinates

Path metadata

  • as_path(): File paths - class "path". Character vectors of image file paths, typically used to track source images for shapes

Measurement data

  • as_meas(): Measurements - class "meas". Numeric vectors of shape measurements (area, perimeter, etc.)

Coefficient classes (all inherit from "coe")

Used for shape descriptors from various decomposition methods:

  • as_coe(): Generic coefficient object - class "coe". Base class for all coefficient types

  • as_eft(): Elliptic Fourier coefficients - class c("eft", "coe"). From closed outline decomposition (outlines)

  • as_rft(): Radii variation Fourier coefficients - class c("rft", "coe"). From radii-based decomposition (outlines)

  • as_dct(): Discrete Cosine Transform coefficients - class c("dct", "coe"). From open curve decomposition (curves)

  • as_npoly(): Natural polynomial coefficients - class c("npoly", "coe"). From polynomial fitting (curves)

  • as_opoly(): Orthogonal polynomial coefficients - class c("opoly", "coe"). From orthogonal polynomial fitting (curves)

  • as_proc(): Procrustes coefficients - class c("proc", "coe"). From Procrustes superimposition (landmarks)

Class hierarchy

coo (coordinates)
├── out (closed outlines)
├── ldk (landmarks)
└── cur (open curves)

coe (coefficients)
├── eft (elliptic Fourier - outlines)
├── rft (radii Fourier - outlines)
├── dct (discrete cosine - curves)
├── npoly (natural polynomial - curves)
├── opoly (orthogonal polynomial - curves)
└── proc (Procrustes - landmarks)

xy (single coordinate matrix)
ldk_id (landmark indices)
path (file paths)
meas (measurements)

See also

declass() to remove Momocs classes

Examples

# Coordinate classes
mat <- matrix(rnorm(100), ncol = 2)
coo <- as_coo(mat)
class(coo)  # "coo" "matrix" "array"
#> [1] "coo"    "matrix" "array" 

outline <- as_out(mat)
class(outline)  # "out" "coo" "matrix" "array"
#> [1] "out"    "coo"    "matrix" "array" 

landmarks <- as_ldk(mat)
class(landmarks)  # "ldk" "coo" "matrix" "array"
#> [1] "ldk"    "coo"    "matrix" "array" 

# Coefficient classes
coefs <- rnorm(24)  # 6 harmonics × 4 coefficients
eft_coefs <- as_eft(coefs)
class(eft_coefs)  # "eft" "coe" "numeric"
#> [1] "eft"     "coe"     "numeric"

# Landmark identifiers (NOT coo)
ldk_indices <- as_ldk_id(c(1, 5, 10, 25))
class(ldk_indices)  # "ldk_id" "integer"
#> [1] "ldk_id"  "numeric"

# In tibbles
library(dplyr)
#> 
#> Attaching package: ‘dplyr’
#> The following objects are masked from ‘package:stats’:
#> 
#>     filter, lag
#> The following objects are masked from ‘package:base’:
#> 
#>     intersect, setdiff, setequal, union
tibble(
  shape = list(mat, mat * 2),
  coef = list(coefs, coefs * 2)
) %>%
  mutate(
    shape = as_out(shape),
    coef = as_eft(coef)
  )
#> # A tibble: 2 × 2
#>   shape    coef 
#>   <out>    <eft>
#> 1 (50 x 2) <24> 
#> 2 (50 x 2) <24>