Geometry Morphometry with the Momocs package

Use ImageMagick processes, in a Python script, to extract flints from standardized photographs:

⬇️

1. Read a XLSX file (download) to recover folder names and photographs filenames
2. Compute several ImageMagick operations (thresholds, connected-components, etc.)
3. Write a black and white image of the lithic with the same filename of the original photo + suffix '_shape'

The black and white JPG image is ready to be used for GMM processes

R Script for the shape analysis and classification of sickles blades

R Script for a demo showing the shape analysis and classification of arrowheads. The original dataset and part of the code comes from:

• Matzig, D. N., Hussain, S. T., & Riede, F. (2021). "Design Space Constraints and the Cultural Taxonomy of European Final Palaeolithic Large Tanged Points: A Comparison of Typological, Landmark-Based and Whole-Outline Geometric Morphometric Approaches". Journal of Paleolithic Archaeology, 4(4), 1-39, doi: https://doi.org/10.5281/zenodo.4560743

R tutorial for butterfly beads.

Alarashi Hala, 2016, "Butterfly beads in the Neolithic Near East: evolution, technology and socio-cultural implications", Cambridge Archaeological Journal

Download the content of the https://github.com/zoometh/Rdev/tree/master/gmm/out folder with https://download-directory.github.io/ and run:

library(Momocs)
library(dplyr)

outpath <- paste0(getwd(), "/gmm/out")

Here, you have to change the outpath to the path of your folder. Start the shape analysis:

lf <- list.files(outpath, full.names = TRUE)
coo <- import_jpg(lf)
beads <- Out(coo)
beads.f <- efourier(beads)

beads is a "Out" "Coo" Momocs' object and beads.f is a "OutCoe" "Coe" Momocs' object (Fourier). Now, show the beads in a panel:

beads %>%
  panel()

Panel

Show them in a standardized stack:

beads %>%
  coo_center %>%
  coo_scale %>%
  coo_slidedirection("up") %T>%
  print() %>%
  stack()

Stack (centered-scaled)

See their clustering

beads.p <- CLUST(beads.f)
plot(beads.p)

Clustering

And show their Kmeans

beads.p <- PCA(beads.f)
KMEANS(beads.p, centers = 4)

Kmeans