I think the difference might be in the indexing: for dip_image objects, the first index goes horizontally, and start at 0. You probably need to switch the x and y coordinates, and might need to add 1 (depending on how you defined the snake earlier).

f = gradient(gradmag(img,30));

with,

[FX, FY] = gradient(double(img));
H = fspecial(‘gaussian’, 30);
FY = imfilter(FY, H);
FX = imfilter(FX, H);

And it works fine, I tried replacing

fex = get_subpixel(f{1},coords,’linear’);
fey = get_subpixel(f{2},coords,’linear’);

with,

F = griddedInterpolant(FX, ‘linear’);
for i = 1:length(x)
fex=F(x,y);
end

G = griddedInterpolant(FY, ‘linear’);
for i = 1:length(y)
fey=G(x,y);
end

etc, and it does something very different.. Is my inderstanding of the sub_pixel function incorrect?

Thanks!

x is a vector, as is fex. Thus, P*(x+gamma*fex) is a matrix times a vector, which results in a vector again. In effect, this statement computes the new coordinates for all snake control points in one operation.

I’m not at all clear with what you’re asking.

fex = get_subpixel(f{1},coords,’linear’);

Is this just returning an intensity at the specified coordinates of the x-grad matrix? When you do,

x = P*(x+gamma*fex);
y = P*(y+gamma*fey);

It looks like you are reassigning the coordinate of the snake to be a matrix (P) multiplied by a constant (x + gamma*fcx), which will mean your coordinates of your matrix on the next pass through will then be matrices? What am I missing?

Thanks

The file ‘bigsur.jpg’ is the stitched result from the first post on panoramic stitching that I wrote (linked in the beginning of this post). If you use that image, the dimensions for the three little images should match and the ‘cat’ command should work fine. But as I said to Prashanth above, this part of the script is only to make the image comparing the three stitching methods, it is not necessary at all for stitching.

Please read the comments in the M-file: the files ‘bigsur1.jpg’ and ‘bigsur2.jpg’ that you find at the top of this post are the downsampled versions of ‘BigSur1.JPG’ and ‘BigSur2.JPG’ that the script looks for. Just skip the first 6 lines of code, and you should be fine!