Can anybody provide some experts's insight in this tough image.
Locked
Can anybody provide some experts's insight in this tough image.
 
|
Hi ImageJ experts,
I did a Micro-CT scanning before using an in-house CT machine (ZEISS Xradia 500 Versa). It shows a strong artefact. I tried many methods to reduce these effects but failed. Can anybody kindly provide some help with this image? Thank you very much. Yongqiang -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
sample 2_tomo-C_recon_Export0501.tiff (2M)
Re: Can anybody provide some experts's insight in this tough image.
|
|
Re: Can anybody provide some experts's insight in this tough image.
|
|
Re: Can anybody provide some experts's insight in this tough image.
|
|
Re: Can anybody provide some experts's insight in this tough image.
|
|
Ok, that means that the scan is "zoomed in" relative to the actual sample dimensions. If there are Fe containing minerals in it, that means that there are some of these minerals outside the edge of the scan; the vertical stripes you see are most likely shadowing effects from these. If you redo your scan zooming out so that the entire sample width is inside the field view, these artifacts should be reduced or disappear completely (since the software would then "see" the mineral grains and be able to correct for it). But you will then also loose resolution; if you have 2µm now on a 2mm sample you will get 5µm on a 5mm sample. Of course that is a major disadvantage. So it seems you are a bit limited by the nature of your sample here.
Regarding accelerating voltage; I usually use minimum 130kV for mineral samples. More energy gives better signal/noise ratio. I do not know the Xradia 500 Versa, perhaps it cannot run at higher voltage for the spotsize required for 2µm resolution? Also, if you increase the number of images to something like 2500 or more (it looks as you are using 1500 or less) the artifacts will be smooth shadows rather than stripes. I can't find much info about the Xradia 500 online but I think it has a CCD detector. Our µCT is a Nikon instrument with a scintillator array and that gives a much better signal noise ratio that what I have seen from CCD detectors. Please see my attached image; it is a 5x5mm crop from a 20x20mm sample imaged at 5 µm resolution. The sample is a composite made of sand with some epoxy filler so it is kinda similar to yours. There are some heavy minerals here too but they do not give any significant artifacts. Of course, comparing a different sample imaged by a different instrument may not help you much. I think your image quality is limited by your sample size and the instrument. Unless you are able to increase the X-Ray energy without losing too much resolution, or increasing the number of images without breaking your budget, there is little you can do. My best suggestion for image processing in ImageJ is to use the FFT → Bandpass Filter to get rid of your artifacts and extract your grains or pores or filler or whatever you want to image and/or analyse. Unfortunately the function does not have a preview, so you need to spend some time finding the best parameters by trial and error. I hope this helps! Stein -----Original Message----- Sent: 12. november 2019 09:09 To: [hidden email] Subject: Re: Can anybody provide some experts's insight in this tough image. Hi Stein, What is your image size, sample size, The resolution of this image is 1.9628 um per pixel. Image size is around 2 mm in diameter. Sample is a cylindrical mini-core plug with diameter of 4.67 mm and length of 14.63 mm drilled from a Berea sandstone plug. It is with a high concentration Fe in the sample mineralogy. acceleration voltage, current, filtering, number of projections, source to detector and source to sample distance? Acceleration voltage is 80 kV. They did not provide the current value and other information. I can only find the above information. Please take a look and Look forward to expert's advice. Thanks, Yongqiang -----Original Message----- From: Stein Rørvik <[hidden email]> Sent: 11 November 2019 16:57 To: [hidden email] Subject: Re: Can anybody provide some experts's insight in this tough image. It appears that you have X-Ray imaging artefacts because your sample is extending outside your field of view. When you do tomography, there must always be clear space at the left and right side of the sample at all rotation angles. Try redoing your scan at a lower magnification (zoom out). It is always better to improve your imaging in the first place than trying to post-correct artefacts on poor quality images! There is also no metadata in this image. What is your image size, sample size, acceleration voltage, current, filtering, number of projections, source to detector and source to sample distance? I could give advice on how to improve the imaging conditions if you provide this. There should be some file in the folder together with the images describing the imaging conditions. What is your material? A porous rock, or a man-made composite? Stein -----Original Message----- Sent: 11. november 2019 11:41 To: [hidden email] Subject: Can anybody provide some experts's insight in this tough image. Hi ImageJ experts, I did a Micro-CT scanning before using an in-house CT machine (ZEISS Xradia 500 Versa). It shows a strong artefact. I tried many methods to reduce these effects but failed. Can anybody kindly provide some help with this image? Thank you very much. Yongqiang -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
| Free forum by Nabble | Edit this page |