RMS image process for multiple photographs

Question

JCH 2020-6-1

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/539519-rms-image-process-for-multiple-photographs

评论： JCH 2020-6-2

I need the RMS result of about 100 pictures like above. First of all, I want to bring the image, use the 'rgb2gray' code, use the 'RMS' code, and add the thresholding value to dichotomize the picture. After all, it is my goal to observe the boundaries between the background and the subject clearly. Please help me with this problem.

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 2020-6-1

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/539519-rms-image-process-for-multiple-photographs#answer_444266

What is the RMS result and RMS code? What does dichotomize mean? You mean binarize or threshold???

What I'd do is, if you want the average of all the spray cone patterns (see the FAQ) is to cast the images to double and sum them up. Then I'd threshold the sum image and call bwboundaries. You might find my interactive thresholding app useful: https://www.mathworks.com/matlabcentral/fileexchange/29372-thresholding-an-image

3 个评论
显示 1更早的评论隐藏 1更早的评论

Image Analyst 2020-6-2

在 MATLAB Online 中打开

OK, easy enough if you have all the S{k}, but how do you get those? I'm sure you have all the raw images but how do they say they "correct" and "denoise" the images before squaring them?

If you just want to find the shape, and not analyze any pattern inside the shape, another option to look at is just to take the max:

for k = 1 : numImages
    outputImageMax = max(cat(3, outputImage, S{k}));
    outputImageRMS = outputImageRMS + double(S{k}) .^ 2;
end
% Need to divide the RMS image (only) by the number of images.
outputImageRMS = outputImageRMS / numImages;

Since it's so easy, do it both ways and look at the images.

JCH 2020-6-2