Resize array, prevent boundary effect

I want to write a script that can double the size of an array using part of the data, while the original array remains the center of the big array. I think selecting some part and flipping it will be a good idea. I wrote the following script, but need assistance on how to assemble the left, right, top, bottom and center of the array to create the double size. I will greatly appreciate help to achieve this.

% read single column data and reshape 
bouin='bou.dat';   % in data
numcolumns=200;    % nx 
numrows=200;       % ny 

fid=fopen(bouin,'r');
bou=fscanf(fid,'%f\n',[numcolumns numrows]);    %open, read in and store
fclose(fid);

% preallocate array with double size of the intial data
bou_doub=zeros(2*numcolumns 2*numrows);

% define the boundary layers to be added
tempa=flipud(bou_doub(1:(numcolumns/2) 1:numrows)); % upper boundary 
tempb=flipud(bou_doub((numcolumns/2):numcolumns 1:numrows)); % lower boundary 
tempc=fliplr(bou_doub(1:numcolumns 1:(numrows/2))); % right boundary 
tempd=fliplr(bou_doub(1:numcolumns (numrows/2):numrows)); % left boundary 

%write all the parts into the main array with double size 
bou_out=bou_doub(

% reshape final array to single column
bou_final=reshape(bou_out(numcolumns*numrows,1));

% write final array to file
fid=fopen('bou_final', 'w');
fprintf(fid,'%6.2f\n',bou_final);  %this opens and writes the output matrix (adding the mean gravity value substracted prior to the FFT) in a ASCII file with a single data column with gravity anomaly
fclose(fid);

[jonas]

Can you also draw the blocks a,b,c and d in your picture?

I guess this row

 %write all the parts into the main array with double size 
bou_out=bou_doub(

is where you are stuck, i.e. you have all the pieces but don't know how to compile them?

[Ope]

thanks Jonas. I tried horzcat for a and b, then vertcat for c and d. I obtained a reasonable result.

Answer 1

[Stephen23]

You could do this quite easily with two very simple indices:

>> A = [0,1,2,3;4,5,6,7;8,9,10,11;12,13,14,15]
A =
    0    1    2    3
    4    5    6    7
    8    9   10   11
   12   13   14   15
>> S = size(A);
>> H = fix(S./2);
>> R = [H(2):-1:1,1:S(2),S(2):-1:H(2)+1]; % row index
>> C = [H(1):-1:1,1:S(1),S(1):-1:H(1)+1]; % column index
>> B = A(C,R)
B =
    5    4    4    5    6    7    7    6
    1    0    0    1    2    3    3    2
    1    0    0    1    2    3    3    2
    5    4    4    5    6    7    7    6
    9    8    8    9   10   11   11   10
   13   12   12   13   14   15   15   14
   13   12   12   13   14   15   15   14
    9    8    8    9   10   11   11   10
>> size(B)
ans =
   8   8

Note that this is quite efficient as it does not copy any of the data into temporary arrays.