nachitox Posted November 16, 2011 Share Posted November 16, 2011 I'll explain with an example: I have this image: http://www.fitnessvital.com/images/galeria/caballo-pastando-0.jpg I create a thumbnail of it: http://www.fitnessvital.com/images/galeria/thumbs/caballo-pastando-0.jpg You can see that it's a big image where the important thing is the horse. The grass is not important. So, the best thing to do it's to crop the extra grass and resizing the rest. So the question is: There's a way of recognizing the objects of an image and crop the rest ? Or identify the non important of an imagen and cropt it ? (the same rewrited) Thank you Quote Link to comment Share on other sites More sharing options...
QuickOldCar Posted November 16, 2011 Share Posted November 16, 2011 This will always be best done manually. Although there are some very complex image programs out there that may be able to do this, I don't believe is a free or usage effective script out there for this. I have seen ways of detecting the most used colors and if is an outer color can crop that, but a computer does not know which part of the image is the most important to you. While I see it may be possible using outlining of objects and group then as the alltogether center image, then use that as a base for center, this is well beyond simple programming. Here are some useful imagemagic scripts you may find useful though. http://www.fmwconcepts.com/imagemagick/innercrop/index.php Quote Link to comment Share on other sites More sharing options...
nachitox Posted November 17, 2011 Author Share Posted November 17, 2011 Yes, its a difficult problem. The imagemagick solution doesnt work for me because the process is automatized and I cant tell whats important (that my goal) I'll ask my machine learning teacher Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.