EECS 442 DIC Project
Useful Links
Overview
This project investigates the application of DIC on images of a brass specimen undergoing a tensile test. These were provided by John Laidlaw, Professional Engineer at the University of Michigan Mechanical Engineering Department.
The goal was to analyze displacement and strain fields in the images to create a stress-strain curve to determine material properties like Young's modulus, Poisson's ratio, ultimate tensile strength, yield strength, and fracture strength. A summary of our findings can be found in our report for EECS 442: Computer Vision course at the University of Michigan. You can view the latest report on Overleaf.
Our source code for analyzing the images can be found in the src folder. The plots generated for the report can be found in the Plots folder. The images of the brass specimen can be found in the Images.
Images and Data
Images of brass specimen undergoing a tensile test were provided by John Laidlaw. The TIFF images were captured with a 5 megapixel Point Grey camera and lens. A LabView program was developed to synchronize the data and image capture throughout the tensile test. John and Dr. Thouless investigated two alternative, open-source software to perform the analysis:
- ncorr (MATLAB)
- GOM Correlate (2D is free)
The data file contains a column with the frame number to sync the data and images. All the images are numbered in sequence (i.e. frame number)
Motivation
The motivation for this project came from Sravan's MECHENG 395: Laboratory I course at the University of Michigan in which a tensile test of brass was conducted and analyzed using LaVision DaVis software. After a conversation with Dr. Michael Thouless, who had worked on a paper using DIC to investigate the cohesive zone in a double-cantilever beam, we determined that this would be a cool project to combine our interest in Mechanical Engineering and Computer Science.
Dependencies
Recommended
Required
- Numpy
- conda install -c anaconda numpy
- Matplotlib
- conda install -c conda-forge matplotlib
- OpenCV
- conda install -c menpo opencv