Unable to perform assignment because the left and right sides have a different number of elements.

1 次查看（过去 30 天）

Jonas Jeskulke 2020-1-7

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/499241-unable-to-perform-assignment-because-the-left-and-right-sides-have-a-different-number-of-elements

评论： Jonas Jeskulke 2020-1-8

problematic_scrip_section.m

I have a problem with the attached script sectrion. anr is a 46x46 double Matrix, with calculated values. I cant find my error but it probably is a very simple one.

Thx for the Help in advance.

10 个评论
显示 8更早的评论隐藏 8更早的评论

Guillaume 2020-1-7

What defines a cluster? Adjacent values (orthogonal? diagonal?) between 0 and 10? If so, have you got the image processing toolbox?

Jonas Jeskulke 2020-1-7

i dont have any toobox s far. I will try to explain: a cluster is formed if the values match ceartain criteria:

1. have to be between >0 and <=10

2. if it is in the same row

3. if it is in another row, but the row it is in is a a collum in an allready exsiting

"cluster matrix" (example: in the first row the second value meets the criteria.

so the values from the second row will be written in the same matrix)

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

Guillaume 2020-1-7

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/499241-unable-to-perform-assignment-because-the-left-and-right-sides-have-a-different-number-of-elements#answer_409027

在 MATLAB Online 中打开

Your criteria for the cluster connectivity is a bit strange in my opinion since it doesn't actually require adjacency. If that's what you want I think the easiest may be to build a graph of connectivity and let matlab split the graph into its components. No toolbox needed for that.

I've tested with the following matrix:

inrange = [1 0 0 1 0 0 1 0 0;
           1 0 1 0 0 0 0 0 0;
           0 1 0 0 1 1 0 0 0;
           0 0 0 0 0 0 1 1 0;
           0 0 0 0 0 0 0 0 1;
           0 1 0 0 1 0 0 0 0];

which according to your criteria has 3 clusters (elements of rows 1, 2, 4 are one cluster, elements of rows 3 and 5 another, and the element of row 6 is another). In your code,

irange = anr > 0 & anr < 10;

With that:

[row, col] = find(inrange);  %locate elements in range
%now build connectivity graph. First iterate over the points to find which points share the same column or row
connectto =  arrayfun(@(idx) find(row(idx) == row | col(idx) == col), 1:numel(row), 'UniformOutput', false);  %Note that this will create self-loop as a point obviously has the same column and row as itself. Self-loops don't matter for connectivity
%convert the above in a 2 column matrix of start and end nodes
connectivity = [repelem((1:numel(row))', cellfun(@numel, connectto)), vertcat(connectto{:})];
%build graph
g = graph(connectivity(:, 1), connectivity(:, 2));
%optionally plot it
plot(g);
%get connected components of the graph, i.e. a cluster by the given criteria of sharing a row or column
clusters = conncomp(g, 'OutputForm', 'cell');
%replace node indices by [row, col] coordinates
clusters = cellfun(@(idx) [row(idx), col(idx)], clusters, 'UniformOutput', false);
%for display:
celldisp(clusters)