Monday, November 9, 2009

most time consuming software

matlab | template matching | thread
In general this can be a very tough problem - numerous people have gotten Ph.D.'s in the subject. Or, if given as homework by a kind professor, it may be crafted to be relatively simple.

Professor is not kind, it is not relatively simply. I've talk with one on my friends who talked with professor and he admitted that 30% efficiency is very good.
The real world will have situations as described in Greg Heath's message, plus other things he didn't mention (but I'm sure knows about) such as noise, partial occlusions (this is a very tough complication, e.g. only part of the apple is visible), orientation changes (e.g. you're trying to find apples and you have the bottom of an apple in your image but your template is the side view of an apple), different shapes/templates (e.g. find tanks given that there are dozens of different models of tanks), color differences (apples come in many different colors and color patterns), etc.

Welcome to real world. Actually I don't understand the exact topic, anyway the desired is to create an algorithm recognizing shapes based on the following pictures and all of them should act both as test and training set:

0 komentarze:

Post a Comment