DIC1

Performs digital image correlation analyses on images obtained experimentally in this scientific article: https://univ-eiffel.hal.science/hal-05115523.

The images were downloaded from Recherche Data Gouv and are distributed under the Etalab Open License 2.0. For further details, see the DATA_LICENSE.md file.

Further implementation details are available in my PhD thesis (Section 2, Chapter 2, written in French).

• 
• 
• 
• 

ROI (691.458 ms)
N_x and N_y (93.371 ms)
Laplacian operator (2.950 ms)
Construct L and M (76.087 ms)

 /img10.png
Iter  1 ||b|| 4.9e-02
Iter  2 ||b|| 1.9e-02
Iter  3 ||b|| 9.4e-03
Iter  4 ||b|| 4.8e-03
Iter  5 ||b|| 2.5e-03
Iter  6 ||b|| 1.3e-03
Iter  7 ||b|| 6.6e-04
Iter  8 ||b|| 3.5e-04
Iter  9 ||b|| 1.8e-04
Iter 10 ||b|| 9.4e-05
Iter 11 ||b|| 4.9e-05
Iter 12 ||b|| 2.6e-05
Iter 13 ||b|| 1.3e-05
Iter 14 ||b|| 7.0e-06
Iter 15 ||b|| 3.7e-06
Iter 16 ||b|| 1.9e-06
Iter 17 ||b|| 1.0e-06
Iter 18 ||b|| 5.3e-07
 /img18.png
Iter  1 ||b|| 7.3e-02
Iter  2 ||b|| 1.2e-02
Iter  3 ||b|| 5.3e-03
Iter  4 ||b|| 2.5e-03
Iter  5 ||b|| 1.2e-03
Iter  6 ||b|| 5.7e-04
Iter  7 ||b|| 2.7e-04
Iter  8 ||b|| 1.3e-04
Iter  9 ||b|| 6.3e-05
Iter 10 ||b|| 3.0e-05
Iter 11 ||b|| 1.5e-05
Iter 12 ||b|| 7.1e-06
Iter 13 ||b|| 3.4e-06
Iter 14 ||b|| 1.7e-06
Iter 15 ||b|| 8.2e-07
 /img26.png
Iter  1 ||b|| 7.1e-02
Iter  2 ||b|| 4.0e-02
Iter  3 ||b|| 1.2e-02
Iter  4 ||b|| 5.4e-03
Iter  5 ||b|| 2.6e-03
Iter  6 ||b|| 1.2e-03
Iter  7 ||b|| 5.9e-04
Iter  8 ||b|| 2.8e-04
Iter  9 ||b|| 1.4e-04
Iter 10 ||b|| 6.7e-05
Iter 11 ||b|| 3.3e-05
Iter 12 ||b|| 1.6e-05
Iter 13 ||b|| 7.8e-06
Iter 14 ||b|| 3.8e-06
Iter 15 ||b|| 1.9e-06
Iter 16 ||b|| 9.2e-07

import matplotlib.pyplot as plt
from PIL import Image  # matplotlib dependency
import numpy as np

from EasyFEA import Display, Folder, Models
from EasyFEA.Simulations import Elastic, DIC
from EasyFEA.Geoms import Circle, Domain


def Plot_Result(simu: Elastic, result: str, img=None, title="", plotMesh=True):
    ax = Display.Init_Axes()
    if img is not None:
        ax.imshow(img, cmap="gray")
    Display.Plot_Result(simu, result, title=title, plotMesh=plotMesh, ax=ax)


if __name__ == "__main__":

    # ----------------------------------------------
    # Config
    # ----------------------------------------------

    useRegularization = True

    imagesDir = Folder.Join(Folder.Dir(), "_images1")

    # ----------------------------------------------
    # Load images
    # ----------------------------------------------

    images = [
        Folder.Join(imagesDir, image)
        for image in sorted(Folder.os.listdir(imagesDir))
        if image.endswith(".png")
    ]

    imgRef = np.asarray(Image.open(images[0]), dtype=int)

    ax = Display.Init_Axes()
    ax.imshow(imgRef, cmap="gray")

    # ----------------------------------------------
    # Create Mesh
    # ----------------------------------------------

    imgScale = 0.11479591836734694  # mm/px

    x0, y0 = 35, 25
    x1, y1 = 423, 814
    meshSize = (x1 - x0) / 20
    contour = Domain((x0, y0), (x1, y1), meshSize)
    contour.Plot(ax)

    xC, yC, radius = DIC.Get_Circle(imgRef, 30.0, [(150, 350), (350, 500)])
    circle = Circle((xC, yC), 2 * radius, meshSize)
    circle.Plot(ax)

    mesh = contour.Mesh_2D([circle])

    Display.Plot_Mesh(mesh, alpha=0, edgecolor="black", ax=ax)
    ax.legend()

    # ----------------------------------------------
    # Conduct DIC analyses
    # ----------------------------------------------

    lr = int(np.sqrt(0.25) * 8 * meshSize) if useRegularization else 0
    dic = DIC(mesh, imgRef, lr, verbosity=True)

    # Generate an elastic simulation to streamline the visualization of results.
    simu = Elastic(mesh, Models.Elastic.Isotropic(2, E=1, v=0.3))

    for image in images[1:]:

        imgStr = image.removeprefix(imagesDir)
        print(f"\n {imgStr}")
        img = np.asarray(Image.open(image), dtype=int)

        u = dic.Solve(img)

        simu._Set_solutions(simu.problemType, u * imgScale)
        simu.Save_Iter()

        if image == images[-1]:
            Plot_Result(simu, "ux", img, f"{imgStr} ux")
            Plot_Result(simu, "uy", img, f"{imgStr} uy")
            Plot_Result(simu, "Exx", img, f"{imgStr} Exx")

    plt.show()