This notebook compares original contours and contours reconstructed from oriented true EFD normalization and true EFD normalization for Quercus serrata and Quercus crispula.
Output figures in this notebook are used in Figure 2 of the manuscript.
Attaching package: 'Momocs'The following object is masked from 'package:stats':
filtersource("R/geometry.R")
source("R/normalization.R")
source("R/compare_contours.R")
# color palette for colorblind-friendly visualization
palette("Okabe-Ito")
# set the data directory from environment variable (.Renviron)
data_dir <- Sys.getenv("PROJECT_DATA_DIR")Load the species names from the CSV file.
This file contains three columns: id, konara, and mizunara. The id column corresponds to the sample ID, while the konara and mizunara columns contain binary values indicating the species classification. We will determine the species name based on which column has a value of 1 for each sample.
konara means Quercus serrata, and mizunara means Quercus crispula. Both are standard Japanese common names used in the original dataset metadata.
Load the contour data from the CSV files in the data/contour_quercus/ directory.
file_paths_xy <- list.files(
file.path(data_dir, "data/contour_quercus/"),
full.names = TRUE,
pattern = "\\.csv$"
)
id_name <- file_path_sans_ext(basename(file_paths_xy))
id_name_head <- lapply(id_name, function(x) {
parts <- unlist(strsplit(x, "_"))
return(parts[1])
})
id_name_head <- unlist(id_name_head)
# match species names to contour data
sp_name <- df_sp_name$species[match(id_name_head, df_sp_name$id)]
# load contour data into a list of matrices
xy_list <- lapply(file_paths_xy, function(fp) {
#cat("File:", fp, "\n")
df <- read.csv(fp)
df <- df[, c("x", "y")]
df <- as.matrix(df)
storage.mode(df) <- "double"
# If the contour is not closed, add the first point to the end to close it
if (!all(df[1, ] == df[nrow(df), ])) {
df <- rbind(df, df[1, ])
}
return(df)
})
names(xy_list) <- id_nameTrue normalized EFD coefficients are obtained by LeafContourEFD.
file_paths <- list.files(
file.path(data_dir, "data/coefficients_efd_normalized_quercus/"),
full.names = TRUE,
pattern = "\\.csv$"
)
ef_list <- lapply(file_paths, function(fp) {
#cat("File:", fp, "\n")
df <- read.csv(fp)
return(df)
})Compare the original contour and the reconstructed contour from oriented true EFD normalization for some samples.
Leftward examples: samples with indices 1, 2, 3, 4, 5, 181, 182, 183, and 184.
idx_left <- c(1:5, 181:184)
labels_left <- letters[1:9]
layout(matrix(seq_along(idx_left), nrow = 3, byrow = TRUE))
for (i in seq_along(idx_left)) {
compare_contour_oriented_true_EFD_normalization(
original = xy_list[[idx_left[i]]],
ef_normalized = ef_list[[idx_left[i]]],
nb.h = 35,
mar = c(0, 0, 0, 0),
lwd = 3,
cex.text = 5,
label = labels_left[i]
)
}
Rightward examples: samples with indices 322, 374, 375, 376, 378, 595, 596, 746, and 372.
idx_right <- c(322, 374, 375, 376, 378, 595, 596, 746, 372)
labels_right <- letters[10:18]
layout(matrix(seq_along(idx_right), nrow = 3, byrow = TRUE))
for (i in seq_along(idx_right)) {
compare_contour_oriented_true_EFD_normalization(
original = xy_list[[idx_right[i]]],
ef_normalized = ef_list[[idx_right[i]]],
nb.h = 35,
lwd = 3,
cex.text = 5,
label = labels_right[i]
)
}
save_dir <- "results"
dir.create(save_dir, showWarnings = FALSE)
save_file_name <- "fig2a_a-i_compare_contours_oriented_true_efd_normalized_quercus_leftward.svg"
svg(file.path(save_dir, save_file_name), bg = "transparent")
idx_left <- c(1:5, 181:184)
labels_left <- letters[1:9]
layout(matrix(seq_along(idx_left), nrow = 3, byrow = TRUE))
for (i in seq_along(idx_left)) {
compare_contour_oriented_true_EFD_normalization(
original = xy_list[[idx_left[i]]],
ef_normalized = ef_list[[idx_left[i]]],
nb.h = 35,
mar = c(0, 0, 0, 0),
lwd = 3,
cex.text = 5,
label = labels_left[i]
)
}
dev.off()save_dir <- "results"
dir.create(save_dir, showWarnings = FALSE)
save_file_name <- "fig2a_j-r_compare_contours_oriented_true_efd_normalized_quercus_rightward.svg"
svg(file.path(save_dir, save_file_name), bg = "transparent")
idx_right <- c(322, 374, 375, 376, 378, 595, 596, 746, 372)
labels_right <- letters[10:18]
layout(matrix(seq_along(idx_right), nrow = 3, byrow = TRUE))
for (i in seq_along(idx_right)) {
compare_contour_oriented_true_EFD_normalization(
original = xy_list[[idx_right[i]]],
ef_normalized = ef_list[[idx_right[i]]],
nb.h = 35,
lwd = 3,
cex.text = 5,
label = labels_right[i]
)
}
dev.off()To assess whether contours reconstructed after true EFD normalization are oriented consistently with the original contours, we randomly rotate the original contours before applying true EFD normalization.
Compare the original contour and the reconstructed contour from true EFD normalization for samples do not match the original orientation (i.e., samples with indices 1, 2, 3, 4, 5, 181, 182, 183, and 184).
idx_left <- c(1:5, 181:184)
labels_left <- letters[1:9]
layout(matrix(seq_along(idx_left), nrow = 3, byrow = TRUE))
for (i in seq_along(idx_left)) {
compare_contour_true_EFD_normalization(
original = xy_list_rotated[[idx_left[i]]],
label = labels_left[i],
nb.h = 35,
mar = c(0, 0, 0, 0),
lwd = 3,
cex.text = 3
)
}
idx_right <- c(322, 374, 375, 376, 378, 595, 596, 746, 372)
labels_right <- letters[10:18]
layout(matrix(seq_along(idx_right), nrow = 3, byrow = TRUE))
for (i in seq_along(idx_right)) {
compare_contour_true_EFD_normalization(
original = xy_list_rotated[[idx_right[i]]],
label = labels_right[i],
nb.h = 35,
mar = c(0, 0, 0, 0),
lwd = 3,
cex.text = 3
)
}
save_dir <- "results"
dir.create(save_dir, showWarnings = FALSE)
save_file_name <- "fig2b_a-i_compare_contours_true_efd_normalized_quercus_leftward.svg"
svg(file.path(save_dir, save_file_name), bg = "transparent")
idx_left <- c(1:5, 181:184)
labels_left <- letters[1:9]
layout(matrix(seq_along(idx_left), nrow = 3, byrow = TRUE))
for (i in seq_along(idx_left)) {
compare_contour_true_EFD_normalization(
original = xy_list_rotated[[idx_left[i]]],
label = labels_left[i],
nb.h = 35,
mar = c(0, 0, 0, 0),
lwd = 3,
cex.text = 3
)
}
dev.off()save_dir <- "results"
dir.create(save_dir, showWarnings = FALSE)
save_file_name <- "fig2c_j-r_compare_contours_true_efd_normalized_quercus_rightward.svg"
svg(file.path(save_dir, save_file_name), bg = "transparent")
idx_right <- c(322, 374, 375, 376, 378, 595, 596, 746, 372)
labels_right <- letters[10:18]
layout(matrix(seq_along(idx_right), nrow = 3, byrow = TRUE))
for (i in seq_along(idx_right)) {
compare_contour_true_EFD_normalization(
original = xy_list_rotated[[idx_right[i]]],
label = labels_right[i],
nb.h = 35,
mar = c(0, 0, 0, 0),
lwd = 3,
cex.text = 3
)
}
dev.off()Visualize all the original contours and the contours reconstructed from oriented true EFD normalization for all samples, and save the plots as SVG files.
list_id <- list(
1:50,
51:100,
101:150,
151:200,
201:250,
251:300,
301:350,
351:400,
401:450,
451:500,
501:550,
551:600,
601:650,
651:700,
701:746
)
for (ids in list_id) {
ids <- unlist(ids)
layout(matrix(1:50, nrow = 5, byrow = TRUE))
for (id in ids) {
compare_contour_true_EFD_normalization(
xy_list_rotated[[id]],
label = names(xy_list_rotated[id]),
nb.h = 35,
lwd = 2,
mar = c(0, 0, 0, 0),
cex.text = 1
)
}
}
save_dir <- "results/supplementary_compare_contours_true_efd_normalized_quercus/"
dir.create(save_dir, recursive = TRUE, showWarnings = FALSE)
list_id <- list(
1:50,
51:100,
101:150,
151:200,
201:250,
251:300,
301:350,
351:400,
401:450,
451:500,
501:550,
551:600,
601:650,
651:700,
701:746
)
for (ids in list_id) {
ids <- unlist(ids)
save_path <- paste0(
save_dir,
"/compare_contours_true_efd_normalized_quercus_",
min(ids),
"_",
max(ids),
".svg"
)
svg(save_path, width = 10, height = 5, bg = "transparent")
layout(matrix(1:50, nrow = 5, byrow = TRUE))
for (id in ids) {
compare_contour_true_EFD_normalization(
original = xy_list_rotated[[id]],
label = names(xy_list_rotated[id]),
nb.h = 35,
lwd = 2,
mar = c(0, 0, 0, 0),
cex.text = 1
)
}
dev.off()
}