Say i have 2 grayscale images.
IMAGE 1: 2 objects
IMAGE 2: 1 object
Now I have worked out the position of the objects.
How could i check in image2 whether any of the objects have been seen in image1?
Say i have 2 grayscale images.
IMAGE 1: 2 objects
IMAGE 2: 1 object
Now I have worked out the position of the objects.
How could i check in image2 whether any of the objects have been seen in image1?
Classify the objects in Image 1, and keep a time/pattern history of the objects seen.
What do you mean by that? What pattern? and by time do you mean position?
Image Classification
Keep track of the objects you have seen in Image 1 and the time they were seen.
After a set time limit discard the records.
For each object seen in Image 2, check the list of objects seen in Image 1 for matches.
How would you use a histogram to compare objects?
Maybe you need associative memory, it is very cool, similar to "handwriten letter recognizing system" that I built...?
Arduino rocks!
Could be interesting, Do you have a simple version I could implement?
You need to fix your signature, more than about 3-5 lines is considered spammy. I's actually so bad it makes your post difficult to read.
If all you know how to use is a hammer, every problem looks like a nail.
That depends on what data the histogram is mapping and how you want to compare the two. I wouldn't suggest a histogram for the problem you posted though, suggest a FIFO queue that keeps track of the objects seen in each picture over time. Keep separate lists for each picture, and compare the two lists to see if an object in picture 1 is latter seen in picture two, or if it was in picture 2 first.
Last edited by abachler; 09-25-2009 at 10:23 PM.
Yes, but with associative memory you could even compare two objects if they were rotated ^^, ANYTHING you want.
Arduino rocks!
COuld you explain how I can do it then Yann please?
perceptrons cannot do that yann. Stop preaching hype that you read on some web page and perhaps pick up a serious text on the subject, here are some of the ones in my library that I found useful -
Artificial Intelligence, A Modern Approach ISBN 0-13-790395-2
Neural Networks, A Comprehensive Foundation, ISBN 0-13-273350-1
Neural Networks for Pattern Recognition, ISBN 978-0-19-853864-6
Biophysics of Computation, Information porcessing in single neurons, ISBN 978-0-19-518199-9
Theoretical Neuroscience, Computational and Mathematical Modelling of Neural Systems, ISBN 978-0-262-54185-5
C++ Neural Networks & Fuzzy Logic, ISBN 1-55851-552-6
Last edited by abachler; 09-26-2009 at 08:32 AM.
so i could still do with a simple basic idea to compare two objects?
You should really post some kind of warning when you include terrible images in your threads.
Quzah.
Hope is the first step on the road to disappointment.
Put a shirt on abachler
You should post your address so we can send you some clothes. Jeez.
Warning: Have doubt in anything I post.
GCC 4.5, Boost 1.40, Code::Blocks 8.02, Ubuntu 9.10 010001000110000101100101