sub2ind error

4 views (last 30 days)
ike
ike on 28 Apr 2012
Commented: Luca Santoro on 30 Apr 2020
i use this to make an accumulator array but it comes error
mat=zeros(250,250);
x=linspace(0,25.1923,100);
y=linspace(-7.9089e+003,313,100);
index=sub2ind(size(mat),round(y),round(x));
i run that code but 'out of range subscript'
that's why ? explain me please

Accepted Answer

Image Analyst
Image Analyst on 28 Apr 2012
You need to convert from "real world units" into units of matrix elements. Below I used the "point-slope" formula for a line to convert your real world units into indexes. Here's one way that's fairly general for linear transforms like you used: (It's not that complicated, just try to follow the comments.)
mat=zeros(250,250);
% Create sample x and y vectors in "real world" units.
x=linspace(0,25.1923,100); % Columns
y=linspace(-7.9089e+003,313,100); % Rows.
% Figure out what indexes those are
% in terms of elements or indexes.
% These indexes will range from 1 to 100 (or the length of x or y).
slopeX = (max(x) - min(x)) / (length(x)-1)
rowIndexes = int32(1 + (x - min(x))/ slopeX)
slopeY= (max(y) - min(y)) / (length(y)-1)
columnIndexes = int32(1 + (y - min(y))/ slopeY)
% Now construct a list of all linear indices in the entire mat matrix.
% (That was wanted for some reason.)
linearIndexes = sub2ind(size(mat), rowIndexes, columnIndexes)
% Now do it for just one particular x and y
% where x and y are in "real world" units.
x0 = 13 % Let's just say you wanted this for example.
y0 = 42
rowIndex = int32(1 + x0 / slopeX)
columnIndex = int32(1 + y0 / slopeX)
% Get the linear index of that one particular point.
linearIndex = sub2ind(size(mat),round(rowIndex),round(columnIndex))
% Report the value of mat() there:
fprintf('mat(%d) = %f\n', linearIndex, mat(linearIndex));
  1 Comment
Luca Santoro
Luca Santoro on 30 Apr 2020
Not work for real negative coordinates

Sign in to comment.

More Answers (2)

Wayne King
Wayne King on 28 Apr 2012
Your x and y vectors are not suitable for sub2ind(). You have an x vector that starts at 0 which is not a valid linear index in MATLAB and you have y-values that are negative, which are not valid linear indices in MATLAB.

ike
ike on 30 Apr 2012
please help me make an accumulator array in hough if the equation y=mx+c

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!