Hi craig, delighted to watch your videos. That's really helpful. May I request you to upload a video how could we normalise an image by using image J software?
Hiii, sir craig! whenever I use Fiji to analyze the particles in my image, I set it to exclude images that are on the edges, but it still considers them. Has the same thing happened to you? Could you help me?
Hello Craig. Thanks for the video, i got very initial idea from this to solve the problem while dealing with SEM image for ceramic material. Could you please tell how can we separate the different objects which in raw image looks like one above or below another so that we can process the image to be used in 2D Analysis. Have a nice day. 😊
Hi Nitin, It's difficult to say without seeing a sample image. My first thought is maybe a watershed process. Threshold your image. Make it binary and then look for the 'watershed' process. If that doesn't work send me a sample image. Find me at University of Glasgow.
Hi Craig. Great job with your videos, very very helpful for me. Just a question, do you have any video for histogram analyses. We are in a research project and need to extract the information for some images for statistic analyses, and the usual histogram extraction is not good. If you can help us would be great.
@@CraigDaly Thanks for the quick response. Yes. We need study the changes in an ultrasound image after and before a Physiotherapy intervention. The problem is that the usual histogram we are extracting is impossible (under our knowledge) to use to be compare with other variables like pain. We have selected the ROI and click on histogram and export to an excel a list of the histogram distribution and must be other method cause in this way the distribution is not detailed and different images can have the same levels of greys. Could you know how?
@@samuelfernandezcarnero9723 hi Samuel. Often with histograms I might group the values into bins of intensity ranges. So maybe divide the data up into 6 or 8 bins. You can then visualise as a bar chart and look for changes in intensity within different ranges. Otherwise, it’s difficult to comment without seeing the images. Send me an example if you want and I might be able to suggest something..
@@samuelfernandezcarnero9723 Hi, i dont like leaving my email address as it can get picked up by bots. Search for me at University of Glasgow. I'm easy to find. C
Nice demonstration Sir..I wanted analyze a SEM image of steel slag grains but the grain edges are clearly not distinguished as slightly it's in contact with adjacent grain ..How do I check it's roundness and circularity I mean shape ..pls help me out..
Hi, without seeing your images I’m guessing at this. Sounds like you need to threshold the image to create roughly circular objects. That binary image can be eroded to hopefully split touching objects. Maybe think of using a watershed filter. Once the objects are split go to analyze particles to get information about the objects. Another option would be to use machine learning. Have a look at my most recent video on Weka Segmentation. Hopefully that helps a bit. Craig.
Hi Craig, thanks for making these great videos! Very useful!
Hi Craig. Awesome job with your videos, it was very helpful for me. Thank you!
This is great - thank you for making this!
Great video ! All of the information are very useful :D
Hi Craig! Thank you very much for the videos!
Great video tutorial - thanks
Awesome video
Hi craig, delighted to watch your videos. That's really helpful. May I request you to upload a video how could we normalise an image by using image J software?
Hiii, sir craig! whenever I use Fiji to analyze the particles in my image, I set it to exclude images that are on the edges, but it still considers them. Has the same thing happened to you? Could you help me?
Hello Craig. Thanks for the video, i got very initial idea from this to solve the problem while dealing with SEM image for ceramic material. Could you please tell how can we separate the different objects which in raw image looks like one above or below another so that we can process the image to be used in 2D Analysis.
Have a nice day. 😊
Hi Nitin, It's difficult to say without seeing a sample image. My first thought is maybe a watershed process. Threshold your image. Make it binary and then look for the 'watershed' process. If that doesn't work send me a sample image. Find me at University of Glasgow.
This tutorial is very useful and easy to operate. But the key problem is how to set threshold and evaluate the mask accuracy.
Yes, good point. Maybe combining with ML/AI will help address these issues in the future.
Hi Craig. Great job with your videos, very very helpful for me.
Just a question, do you have any video for histogram analyses. We are in a research project and need to extract the information for some images for statistic analyses, and the usual histogram extraction is not good. If you can help us would be great.
Hi, thanks for the kind comment. Are you obtaining an image histogram and the exporting the numbers to a spreadsheet?
@@CraigDaly Thanks for the quick response.
Yes. We need study the changes in an ultrasound image after and before a Physiotherapy intervention. The problem is that the usual histogram we are extracting is impossible (under our knowledge) to use to be compare with other variables like pain. We have selected the ROI and click on histogram and export to an excel a list of the histogram distribution and must be other method cause in this way the distribution is not detailed and different images can have the same levels of greys. Could you know how?
@@samuelfernandezcarnero9723 hi Samuel. Often with histograms I might group the values into bins of intensity ranges. So maybe divide the data up into 6 or 8 bins. You can then visualise as a bar chart and look for changes in intensity within different ranges. Otherwise, it’s difficult to comment without seeing the images. Send me an example if you want and I might be able to suggest something..
@@CraigDaly ok. Let me know an email and will send the example if you want. Many thanks for the help.
@@samuelfernandezcarnero9723 Hi, i dont like leaving my email address as it can get picked up by bots. Search for me at University of Glasgow. I'm easy to find. C
Nice demonstration Sir..I wanted analyze a SEM image of steel slag grains but the grain edges are clearly not distinguished as slightly it's in contact with adjacent grain ..How do I check it's roundness and circularity I mean shape ..pls help me out..
Hi, without seeing your images I’m guessing at this. Sounds like you need to threshold the image to create roughly circular objects. That binary image can be eroded to hopefully split touching objects. Maybe think of using a watershed filter. Once the objects are split go to analyze particles to get information about the objects. Another option would be to use machine learning. Have a look at my most recent video on Weka Segmentation. Hopefully that helps a bit. Craig.
nice