Prev: how to make available a variable to two callbacks?
Next: how to make SVM toolbox functioning?
From: Kevin on 21 Mar 2010 11:31
I am not sure if the title above is approriate for my problem.
Basically I have an image posted below, where I have applied mixture of gaussian
and obtained the resultant image although it is still work in progress to make the
the foreground (peoples) more defined but what I want is to scan the image from left to right and identify the small 'white' pixels in the background and remove them so that I simply have only the people and the cars.

http://img175.imageshack.us/img175/3533/screenhunter02mar211413.gif

Any advice would be greatly appreciated.

Thanks
From: ImageAnalyst on 21 Mar 2010 11:43
Kevin
Use bwareaopen(). You could also use the median filter, although this
has less accuracy in removing blobs of a certain size.
From: Kevin on 21 Mar 2010 12:12
ImageAnalyst <imageanalyst(a)mailinator.com> wrote in message <d0dd4208-c691-4f10-b79f-79a2272fb8f7(a)q15g2000yqj.googlegroups.com>...
> Kevin
> Use bwareaopen(). You could also use the median filter, although this
> has less accuracy in removing blobs of a certain size.

Thanks 'ImageAnalyst'. When you say less accuracy does it mean the people and cars are not well defined???

Cheers
From: Kevin on 21 Mar 2010 12:20
ImageAnalyst used the bwareaopen() function and yes it work. But main issue since the foreground is not so well defined that even the people slowly disappear.
From: ImageAnalyst on 21 Mar 2010 13:22
Kevin:
The bwareaopen() function considers only the area and pixels of the
blob itself so of course it's very accurate. The median filter
considers the value of ANY pixels that are contained in the window
whether they belong to the blob of the center pixel, or to any other
blob that's not even connected to the center pixel. Thus the median
filter will give you inaccurate results as the window size gets larger
and starts to include more unconnected blobs.

I can't help you with the larger, more complicated problem of
accurately outlining people and cars in any outdoor scene. This is a
complicated situation worthy of a Ph.D. (unless you have very rigid
constrained tightly controlled situations). Presumably that's what
you're working on, so good luck with that.

Perhaps this link of virtually all the image processing papers ever
published will be of use:
http://iris.usc.edu/Vision-Notes/bibliography/contents.html
 |  Next  |  Last
Pages: 1 2 3
Prev: how to make available a variable to two callbacks?
Next: how to make SVM toolbox functioning?