How to fix error: "Index in position 1 is invalid. Array indices must be positive integers or logical values" caused by bad kernel

Hello everyone! I am making an automatic software in MATLAB (R2019b) which converts 2D images (e.g. depth maps, reflective maps etc.) into a 3D point cloud matrix. I have been investigating methods of denoising which finds the neighbours of a single point in a rxr kernel and stores their indices. This process is repeated for all points. I can't seem to find a reliable method of scanning the points on the edges as it will return an error that reads "Index in position 1 is invalid. Array indices must be positive integers or logical values." I thought padding the image in zeros would help but that seems to be only a temporary solution because once I increase the radius of the kernel, the same error pops up. If anyone can help solve this I would really appreciate it! I'll attach the code below:

clc; clear all; close all;
% Image and radius are going to be function variables but for testing purposes are set to constants
Image = [21:2:37; 90:98; 31:2:47; 71:79; 53:2:69; 41:49; 21:29; 81:89; 11:19]  %input matrix
Radius = 3; % function variable
r = Radius - 2; % box filter parameter
%Indices of neighbouring points in a moving kernel for point cloud
Image = padarray(Image,[r,r],0,'both')
n = 1; % enables for indice loop
[row,col] = size(Image);


 for i = 2: row - 1
        for j = 2: col -1
          LocationMat=zeros(row,col)
          LocationMat(i-r:i+r,j-r:j+r)= 1 %Kernel rxr
          Indices(n,:)=find(LocationMat);   %finds the indices of the box kernel
          n = n + 1;   
        end
        
 end

I need some suggestions on a better method of implementing a rxr kernel since the kernel doesn't work anything but for 3x3. The problem is I cant the edges of the image so i cant get the neighbours of the edges.

Image =

    21    23    25    27    29    31    33    35    37
    90    91    92    93    94    95    96    97    98
    31    33    35    37    39    41    43    45    47
    71    72    73    74    75    76    77    78    79
    53    55    57    59    61    63    65    67    69
    41    42    43    44    45    46    47    48    49
    21    22    23    24    25    26    27    28    29
    81    82    83    84    85    86    87    88    89
    11    12    13    14    15    16    17    18    19
    
    
LocationMat =

     @     1     0     0     0     0     0     0     0     0     0
     1     1     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0     0     0

I want to have a method of getting the neighbours for the @ point, by maybe a method of replicating the matrix when the edge is detected/

Answer 1

[Star Strider]

I am not certain what you are doing.

One option may be the imfilter function.

Another option would be to set ‘hard’ limits on the indices, for example:

LocationMat(max(1,i-r):min(i+r,size(Image,1)), max(1,j-r):min(j+r,size(Image,2)))) = 1 %Kernel rx

I cannot test this because I am not certain what you are doing, so I defer to you to experiment with it.