I really Need suggestions for the next. (Image processing)

[angoav98]

I have very images. These images form  a total  image then I Require to unite all image for complete the total image. I think build the total image with a matrix store the image in the position, but I have a problem: The pictures have a common regions, then if I use the array I will have the same region in two parts of total image, then I need join the images such that the common region of images only appear once.. The common region have 80 pixel.

Then I have very images and need build with these image one total image, but these image have common region then in the total image  appear one part more once

[Hornless.Rhino]

Is the overlap always 80 pixels and is it always there?

If so, here's one way of doing it....

1) Convert images to arrays

2) Strip unnecessary pixels

3) join arrays into  one big array

4) convert array back to an image

There are VIs for doing 1 (image to array), 3 and 4 (array to image) already.

So the issue is step 2.

Lets look at a simple example.

Imagine you have the following group of arrays that form a 3x1 grid. (left to right)

11111

12222

12334

12222

11111

11111

22222

34443

22222

11111

11111

22221

43321

22221

11111

Now imagine you know that there is an overlap of 2 columns between neighbouring arrays.

You could do one of the following:

remove 2 columns on each side of the second array

remove 2 colums from the right of the first array and the left of the third array

remove 2 columns from the right of the first and second array

remove 2 colums from the left of the second and third array

Lets assume you do the first one.

You'd then have the following arrays:

11111

12222

12334

12222

11111

1

2

4

2

1

11111

22221

43321

22221

11111

Joining them together would then give you your un-region duplicated array.

11111111111

12222222221

12334443321

12222222221

11111111111