Prev: pdepe: discontinuous source
Next: Rayleigh Fading
From: ImageAnalyst on 17 Jan 2010 00:25
Sorry - no. Doesn't look like anything I'm familiar with, especially
with no context. Who told you that? Ask them for a more detailed
explanation if you want one.
From: Shamim on 17 Jan 2010 14:10
ImageAnalyst <imageanalyst(a)mailinator.com> wrote in message <7bba457b-6cf5-4ea7-a706-f316d92104f4(a)u41g2000yqe.googlegroups.com>...
> Sorry - no. Doesn't look like anything I'm familiar with, especially
> with no context. Who told you that? Ask them for a more detailed
> explanation if you want one.

What I'm gonna do exactely is to insert the raw histogram bins, of different types and dimensions, as input into my SVM classifier.But I don't understand the concept of RGB(3*16) and RGB(16*16*16) and their difference, the same as HSI(3*16 and HSI(16*16*16)!!
And here is what I have:
>>I =imread(Im.jpg) %original image (1944*2592*3 unit8)
x=[x1,x2,x3,x4]; y=[y1,y2,y3,y4]; % contour coordinates
>>Contour =double((roipoly(I, str(i).x, str(i).y)) %str is the structure containing coordinates of image (1944*2592 logical)
From: ImageAnalyst on 17 Jan 2010 14:45
Perhaps they're talking about histograms where you bin the full
intensity image into bins of size 16, and you can either take the
three histograms of the color channels each one at a time, so that you
have three histograms, each one being 16 elements long, or they
calculate the actual 3D histogram so that you have a 3D array 16 bins
along the red dimension, 16 bins along the green dimension, and
finally 16 bins along the third, blue, dimension. Two different ways
of using the histograms of a color image.

Still not sure what you want to do though. Why don't you give more
complete code? Make sure it's commented exceptionally well, with
copious comments and nice long explicit descriptive variable names.
And post your image to http://drop.io
From: Shamim on 17 Jan 2010 15:26
ImageAnalyst <imageanalyst(a)mailinator.com> wrote in message <7bba457b-6cf5-4ea7-a706-f316d92104f4(a)u41g2000yqe.googlegroups.com>...
> Sorry - no. Doesn't look like anything I'm familiar with, especially
> with no context. Who told you that? Ask them for a more detailed
> explanation if you want one.

What I'm gonna do exactely is to insert the raw histogram bins, of different types and dimmensions, as inputs into my SVM classifier. But I don't understand the concept of RGB(3*16) and RGB(16*16*16) and their difference, as well as HSI(3*16) and HSI(16*16*16)!

and here's what I have:
>>I =imread(Im.jpg) %original image(1944*2592*3 unit8)
x=[x1,x2,x3,x4]; y=[y1,y2,y3,y4]; %contour coordinates
>>contour = roipoly(I,str(i).x, str(i).y)); %str is the structure containing contour coordinates (1944*2592 logical)
From: ImageAnalyst on 17 Jan 2010 16:14
Yes, I saw it the first time you posted it a couple of hours ago. And
I responded to that one. Did you see it? No need to post your reply
a third time. Do you have anything NEW to add?
First  |  Prev  |  Next  |  Last
Pages: 1 2 3
Prev: pdepe: discontinuous source
Next: Rayleigh Fading