Am not able to convert set of image coordinates into training set

plz see attachment herewith for my problem.

[Merged from Attachment]

First of all a set of n images I have obtained. This can be represented as the following set.

A=[I1,I2,I3,I4……………….Im].

Each image has a set of coordinates which represent the pictures. Each coordinate is represented by (x,y) pair denoting the position in the image. M represents the number of images in the training set. The matrix is as follow:

But am not able to convert the set of coordinates into a training set, which is represented as follows: Please help for sorting it out…………. thanks

[Walter Roberson]

Many of the volunteers will not download .zip, so I have extracted the information and posted it here, like you could have done.

[Walter Roberson]

What data type is your veinmat? Is it a cell array in which each entry is a row vector of two values?

[Balaji M. Sontakke]

yes sir, it is a cell array, actually it requires for reducing the variation in the element of the training set and to make it standardize.

Answer 1

[Walter Roberson]

T = zeros(size(veinmat) .* [2 1]);
T(1:2:end,:) = cellfun(@(V) V(1), veinmat);
T(2:2:end,:) = cellfun(@(V) V(2), veinmat);

[Balaji M. Sontakke]

I would like to calculate mean, I have used following statement

mean_a = mean(cellfun(@mean, T,'UniformOutput',0))

But it produces error like...

Error using cellfun
Input #2 expected to be a cell array, was double instead.

[Walter Roberson]

mean_a = mean(mean(T))

[Balaji M. Sontakke]

thanks sir...