How to get the bright areas separately from an image?

Question

Hao Shi 2021-2-20

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/751469-how-to-get-the-bright-areas-separately-from-an-image

评论： Hao Shi 2021-2-25

Hi there,

I have a combustion image in gray scale (Sample), with this I would like to extract the bright areas separately (Sample02).

However, the areas are connected with other areas (merge together), I could not extract them out from the whole area after binarizing the image.

PS: the bright areas are resulted from auto-ignition and the corresponding area and intensity go up very fast, so I want to extract them for analysis.

Any possible solution will be highly appreciated, thanks a lot!

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Image Analyst 2021-2-21

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/751469-how-to-get-the-bright-areas-separately-from-an-image#answer_628879

在 MATLAB Online 中打开

Can you divide the later image by an earlier one? The ratio in the regions where the intensity went up faster than the rest of the image will have a ratio bigger than the rest of the image. Then you can threshold and call regionprops().

ratioImage = double(image2) ./ double(image1);
mask = ratioImage > someValue;
props = regionprops(mask, image2, 'Area', 'MeanIntensity')

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Hao Shi 2021-2-25

Dear Analyst,

Thank you so much for your kind suggestion, it works!!

I tried this method before, there were some noises along the flame boundary, where the intensities also change significantly when the flame propagates. Later I used the filter to filter out the noises, and it looks ok. Thanks a lot!

Before filtering: