how to do logarithmic image normalization?

Question

mahesh chathuranga 2013-10-13

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/90055-how-to-do-logarithmic-image-normalization

评论： Adam Nustian 2016-6-19

采纳的回答： Image Analyst

what is logarithmic image normalization?how to do it?(range is 0-255)

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 2013-10-13

3
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/90055-how-to-do-logarithmic-image-normalization#answer_99503

编辑：Image Analyst 2013-10-15

在 MATLAB Online 中打开

I suppose you just get a new image where the image is the log of the pixel value. It's used to enhance dark areas by expanding their range, while not clipping bright areas. Run the code below for a demo.

workspace;  % Make sure the workspace panel is showing.
% Read in and display the original gray scale image.
grayImage = imread('Cameraman.tif');
grayImage = double(grayImage);
subplot(2,1,1);
imshow(grayImage, []);
axis on;
title('Original Image', 'FontSize', 15);
% Enlarge figure to full screen.
set(gcf, 'Units', 'Normalized', 'OuterPosition', [0 0 1 1]);
% Give a name to the title bar.
set(gcf, 'Name', 'Demo by ImageAnalyst', 'NumberTitle', 'Off') 
% Take the log of it.  Add 1 to avoid taking log of zero.
logImage = log(grayImage+1);
% Normalize to the range 0-1.
normalizedImage = mat2gray(logImage);
% Display it.
subplot(2,1,2);
imshow(normalizedImage, []);
axis on;
title('Log Image', 'FontSize', 15);
% If you want a uint8 version, then you can multiply by 255
% uint8Image = uint8(255 * normalizedImage);
msgbox('Note how the coat has more details');

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Adam Nustian 2016-6-19

ImageAnalyst Are you available to review a paper on Log. Image Processing for an international conference?

Thanks

请先登录，再进行评论。

Answer 2

Anand 2013-10-14

2
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/90055-how-to-do-logarithmic-image-normalization#answer_99580

在 MATLAB Online 中打开

It sounds like you want to just take the log of the image and rescale the range from 0 to 255. This is something you might want to do to visualize a frequency domain signal. Here's how you can do that:

% take logarithm of image.
imLog = log(im);
% normalize
minLog = min(imLog(:));
maxLog = max(imLog(:));
imLogNorm = 255 * (imLog - minLog)./(maxLog-minLog);