Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

Thread Subject:
select features

Subject: select features

From: blue ice

Date: 27 Jul, 2010 13:23:05

Message: 1 of 6

i have extracted features from an image i-e SIFT features from this link http://www.cs.ubc.ca/~lowe/keypoints/

now i have features in a matrix.
now i want to Firstly Select top most feature (at thr right side of image) and bottom most feature (at the right side of image).
than draw a line between these two features.

can you tell me how to do this.

Subject: select features

From: us

Date: 27 Jul, 2010 13:34:20

Message: 2 of 6

"blue ice" <chase_the_dawn_2000@yahoo.com> wrote in message <i2mmjp$4bd$1@fred.mathworks.com>...
> i have extracted features from an image i-e SIFT features from this link http://www.cs.ubc.ca/~lowe/keypoints/
>
> now i have features in a matrix.
> now i want to Firstly Select top most feature (at thr right side of image) and bottom most feature (at the right side of image).
> than draw a line between these two features.
>
> can you tell me how to do this.

this is too vague(!)...
you need to provide more information in ML language, eg,
- small sample of your input...
- an idea of how your output should look like...
- maybe(!) even a snippet of ML code...

us

Subject: select features

From: blue ice

Date: 27 Jul, 2010 14:07:19

Message: 3 of 6

"us " <us@neurol.unizh.ch> wrote in message <i2mn8s$hrf$1@fred.mathworks.com>...
> "blue ice" <chase_the_dawn_2000@yahoo.com> wrote in message <i2mmjp$4bd$1@fred.mathworks.com>...
> > i have extracted features from an image i-e SIFT features from this link http://www.cs.ubc.ca/~lowe/keypoints/
> >
> > now i have features in a matrix.
> > now i want to Firstly Select top most feature (at thr right side of image) and bottom most feature (at the right side of image).
> > than draw a line between these two features.
> >
> > can you tell me how to do this.
>
> this is too vague(!)...
> you need to provide more information in ML language, eg,
> - small sample of your input...
> - an idea of how your output should look like...
> - maybe(!) even a snippet of ML code...
>
> us

i am using this

[image1, descriptors1, locs1] = sift(image1);

so i have this data
image1: the image array in double format
descriptors1: a K-by-128 matrix, where each row gives an invariant descriptor for one of the K keypoints. The descriptor is a vector of 128 values normalized to unit length.
locs1: K-by-4 matrix, in which each row has the 4 values for keypoint location (row,column, scale, orientation). The orientation is in the range [-PI, PI] radians.

i want to get two descriptors,one located at top of image and other at the bottom of the image (but at right side).
than draw a line between these two.

hope u get my point :(

Subject: select features

From: us

Date: 27 Jul, 2010 14:20:20

Message: 4 of 6

"blue ice" <chase_the_dawn_2000@yahoo.com> wrote in message <i2mp6n$pvk$1@fred.mathworks.com>...
> "us " <us@neurol.unizh.ch> wrote in message <i2mn8s$hrf$1@fred.mathworks.com>...
> > "blue ice" <chase_the_dawn_2000@yahoo.com> wrote in message <i2mmjp$4bd$1@fred.mathworks.com>...
> > > i have extracted features from an image i-e SIFT features from this link http://www.cs.ubc.ca/~lowe/keypoints/
> > >
> > > now i have features in a matrix.
> > > now i want to Firstly Select top most feature (at thr right side of image) and bottom most feature (at the right side of image).
> > > than draw a line between these two features.
> > >
> > > can you tell me how to do this.
> >
> > this is too vague(!)...
> > you need to provide more information in ML language, eg,
> > - small sample of your input...
> > - an idea of how your output should look like...
> > - maybe(!) even a snippet of ML code...
> >
> > us
>
> i am using this
>
> [image1, descriptors1, locs1] = sift(image1);
>
> so i have this data
> image1: the image array in double format
> descriptors1: a K-by-128 matrix, where each row gives an invariant descriptor for one of the K keypoints. The descriptor is a vector of 128 values normalized to unit length.
> locs1: K-by-4 matrix, in which each row has the 4 values for keypoint location (row,column, scale, orientation). The orientation is in the range [-PI, PI] radians.
>
> i want to get two descriptors,one located at top of image and other at the bottom of the image (but at right side).
> than draw a line between these two.
>
> hope u get my point :(

no...
in this NG you MUST produce something that CSSMers can PLAY with, eg,
a snippet of input data/desired output...

sorry
us

Subject: select features

From: blue ice

Date: 27 Jul, 2010 16:05:03

Message: 5 of 6

What do you want to say????is this a bad approach or is it not possible to do?

Subject: select features

From: Steven_Lord

Date: 27 Jul, 2010 16:32:22

Message: 6 of 6



"blue ice" <chase_the_dawn_2000@yahoo.com> wrote in message
news:i2n03f$6gr$1@fred.mathworks.com...
> What do you want to say????is this a bad approach or is it not possible to
> do?

us is saying that you haven't given him enough information to be able to
answer.

Put a sample image (or a few) on some free file hosting site and post a link
to where those images are located, and explain SPECIFICALLY the results you
want to receive when you operate on those sample images (and as much about
how you want to obtain those results as you can.) That should be sufficient
information to allow the gurus to at least give a first pass approach to
what you want to do (or to ask more targeted questions about what they don't
understand in your explanation.)

--
Steve Lord
slord@mathworks.com
comp.soft-sys.matlab (CSSM) FAQ: http://matlabwiki.mathworks.com/MATLAB_FAQ
To contact Technical Support use the Contact Us link on
http://www.mathworks.com

Tags for this Thread

No tags are associated with this thread.

What are tags?

A tag is like a keyword or category label associated with each thread. Tags make it easier for you to find threads of interest.

Anyone can tag a thread. Tags are public and visible to everyone.

Contact us