Issue Plotting Array in Imagesc

Question

Rose 2024-3-5

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2090391-issue-plotting-array-in-imagesc

评论： Jatin 2024-8-8

I am trying to plot an array as an imagesc plot. The array is full of non-zero numbers, but imagesc plots them as zero. I think the issue is with scaling, but I don't know how to fix it. Screenshots attached to show what I mean. How do I scale my imagesc plot so that the numbers in the array are visible appropriately? Tried looking it up on the documentation but couldn't find a clear answer for my problem.

3 个评论
显示 1更早的评论隐藏 1更早的评论

DGM 2024-3-5

It might be worth looking at a histogram of the data to see if it would be prudent to pick colorscale limits other than the data extrema.

Mathieu NOE 2024-3-5

hello @Rose

it would help if you could share your data (and code if you have one)

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Jatin 2024-8-7

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2090391-issue-plotting-array-in-imagesc#answer_1495874

在 MATLAB Online 中打开

Hi,

This happens when “imagesc” automatically tries to scale the colormap based on the range of data in array. It makes it harder to visualise large range data using colormap if there are few outliers.

Few things that you might want to try are:

Remove outliers from the data, this will make the visualisation of data better using automatic colormap scaling.
Manually set the colormap limits using the “clim” property of the “imagesc” function, which allows you to define the range of data values that map to the colours.

For e.g.,

data = rand(10, 10) * 100; 
% Create the imagesc plot
figure;
imagesc(data);
colorbar; 
% Set the colour limits to the min and max of the data
clim([min(data(:)), max(data(:))]);

Kindly refer to “clim” documentation using link mentioned below:https://www.mathworks.com/help/releases/R2023b/matlab/ref/clim.html

2 个评论
显示无隐藏无

DGM 2024-8-8

在 MATLAB Online 中打开

Your example does the opposite of what your text describes. Using imagesc() already sets CDataMapping to 'scaled' and sets clim to the data extrema by default.

As you say, you either need to remove outliers, or simply pick different limits for clim().

% a bunch of data close to zero, but with very wide tails

A = 1E6./randn(1000);

histogram(A)

% default scaling

imagesc(A)

% pick better limits

pct = 10;

newrange = prctile(A,[pct 100-pct],'all');

imagesc(A)

clim(newrange)

Jatin 2024-8-8

Hi @DGM, I stated the example just to show how "clim" can be used, but you are right I could have used a better limit. Thanks for your informative example.

请先登录，再进行评论。