Legacy ImageMagick Discussions Archive

Hi all, I would like to know if there is any way to detect the coordinates of several images inside another (bigger) one.

The actual situation is this:
- I have several JPEG images, all of them with a WHITE background and several little images over that white background
- I would like to obtain the position of the squares which could cover each image. I mean, if one of those little images is a triangle, i would like to obtain the coordinates of the smaller square which could cover entirely that triangle.
- My intention is to locate each little image with something similar to this information: coordinate_X + coordinate_Y + width + height ...or maybe something like this... coordinateA + coordB + coordC + coordD <-- Both cases let me to define the squares I am looking for.

So, the result for 1 big image detection could be something like this:
Detected little image #1: (100,100,50,50) <-- The little image is placed in the position 100, 100 and its size is 50px X 50px
Detected little image #2: (500, 150, 75, 30)
Detected little image #3: (600, 300, 40, 100)

Do you know any way to get what I want to obtain?? Or maybe another new compatible idea !!! I'm open minded and I really don't know how to manage this

Thank yo very much!!

I don't know if these will help but they do similar kinds of things. See my bash unix schell scripts, multicrop and separate at the link below. The virtual-canvase (geometry information) in the verbose info: for each image may contain the bounding box coordinates for that image, but I am not sure without looking more closely.

Thank you very much, fmw42.
I'll put an eye on them in order to discover if I have enogh knowledgements to mix them to get what I need.

Bye!