Please, does anyone have any idea about the message I posted earlier
(below).

Regards
Betty

On Tue, Mar 20, 2012 at 3:04 AM, Betty Desire <>wrote:

> Hello,
>
> Does anyone know how to use the SVM in weka to do classification on
> images. See an example attached. It is a processed road image. I couldn't
> attach the original frame due to the constraint below. The first mail I
> sent bounced back.
>
> Message body is too big: 100557 bytes with a limit of 40 KB
>
> Any help is appreciated.
>
> Regards
> Betty
>

On 23/03/12 5:45 PM, Betty Desire wrote:
> Please, does anyone have any idea about the message I posted earlier
> (below).
>
> Regards
> Betty
>
> On Tue, Mar 20, 2012 at 3:04 AM, Betty Desire <
> > wrote:
>
> Hello,
>
> Does anyone know how to use the SVM in weka to do classification on
> images. See an example attached. It is a processed road image. I
> couldn't attach the original frame due to the constraint below. The
> first mail I sent bounced back.

You will have to extract features from your images and represent each as
a feature vector first. Weka doesn't have any built-in image
processing/extraction tools. You could take a look at FeatureJ:

http://www.imagescience.org/meijering/software/featurej/

Cheers,
Mark.

_______________________________________________
Wekalist mailing list
Send posts to:
List info and subscription status: https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
)

Hello,
>>
>> Does anyone know how to use the SVM in weka to do classification on
>> images. See an example attached. It is a processed road image. I
>> couldn't attach the original frame due to the constraint below. The
>> first mail I sent bounced back.
>>
>
> You will have to extract features from your images and represent each as a
> feature vector first. Weka doesn't have any built-in image
> processing/extraction tools. You could take a look at FeatureJ:
>
> http://www.imagescience.org/meijering/software/featurej/
>
> Cheers,
> Mark.
>
> _______________________________________________
> Wekalist mailing list
> Send posts to:
> List info and subscription status:
> https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette:
> http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
>

Thanks Mark for the response,

What format does the feature matrix/vector have to be in order to be able
to use the SVM in weka. Do I need to merge all feature vectors for all
training images or separately for each image. I mean getting the training
set and the test set. Your help is much appreciated.

Regards
Betty

On 29/03/12 5:57 PM, Betty Desire wrote:
>
>
> Hello,
>
> Does anyone know how to use the SVM in weka to do
> classification on
> images. See an example attached. It is a processed road image. I
> couldn't attach the original frame due to the constraint
> below. The
> first mail I sent bounced back.
>
>
> You will have to extract features from your images and represent
> each as a feature vector first. Weka doesn't have any built-in image
> processing/extraction tools. You could take a look at FeatureJ:
>
> http://www.imagescience.org/meijering/software/featurej/
>
> Cheers,
> Mark.
>
> _______________________________________________
> Wekalist mailing list
> Send posts to:
>
> List info and subscription status:
> https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette:
> http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
>
>
>
> Thanks Mark for the response,
>
> What format does the feature matrix/vector have to be in order to be
> able to use the SVM in weka. Do I need to merge all feature vectors for
> all training images or separately for each image. I mean getting the
> training set and the test set. Your help is much appreciated.

Each image should end up as a separate instance in either the training
or testing ARFF file.

Cheers,
Mark.

_______________________________________________
Wekalist mailing list
Send posts to:
List info and subscription status: https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
)