Hello,
Can someone with experience in CT image analysis help me to understand the elements of the pictures. It is a plate thermoformed from waste like ABS and PMMA polymers and fiberglass. I need to know what black dots represent, white particles, and also if fiberglass is visible.
Thank you!
This would be a great question to post in our new free medical imaging question and answer forum ( www.imagingQA.com ), there are a number of CT and image processing experts in the community. If useful, please feel free to open a new topic at the link below :
https://community.imagingqa.com/c/questions/5
Try to find the a table with hounsfield units of different materials. A quick search found me two here on researchgate.
However, it depends on the materials you have in the plate and (with eyeballing only) on the windowing of your images. Better to measure the HU in the images. for example PMMA is around -300 HU, thus it should be black (water is per definitionem 0 HU).
As previous replies have said, the best way to determine different materials would be to examine the Hounsfield Units (HU) on the image.
Generally speaking, for most CT processing kernels, the less dense a material is, the lower the HU will be and the darker (more black) it will appear on a CT image.
However, the exact range of Hounsfield Units for a given material will also depend on the parameters (tube voltage, current) that were used for the scan acquisition.
A quick search yielded this paper which examines the HU for a number of different 3D printed plastics: Article Tissue-equivalence of 3D-printed plastics for medical phanto...
I think in this case the best way to determine what different materials are present is to also perform control scans of samples of each individual material, this will allow you to determine the HU for each of them and serve as a starting point for analyzing your mixed medium samples.
Dear Mihai, I can see that you are using our Software from Volume Graphics. The best would be to go through the training materials first and get an understanding of CT data. What you see is always the different absorption of X-Rays of different materials. The darker the "dots", the closer the absorptin is to the back groud material. In fact it can also be porosity of it is very similar. The "lighter" the dots are, the more absorption there is for X-Rays. It does not necessarily match the houndsfield scale at all, because industrial CT's do not get calibrated for medical purposes and then water is not having a houndsfield value of zero and air of 1000 what a doctor may expect. The grey values may reach to several thousands. If you can see fibers actually depends on the resulution you get from your setup. Check for the voxel size (properties of the dataset, rightclick in the scene tree). The voxel size needs to get very close to the fiber size, otherwise it won't show up as it cannot be resolved. Best get some infos from our support center. I think we can help with some basics in the physics and a training in the software.
The other answers are all reasonable, and you should educate yourself on CT. However, another way to approach this is to image the component materials with known distributions (a block of each material after undergoing the same heat treatment). If the thermaforming doesn't change the electron density of the material, then they could be imaged in the pre-combo state. The resultant Hounsfield units can then be mapped to your image, and using ImageJ or your favorite tool, you can then segment based on HU range to quantify. That would be a fairly elegant first step.
The larger the gray value in the figure, the denser the composition of the material, the black things are likely to be poresthe figure, the denser the composition of the material, the black things are likely to be pores