[scikit-learn] Regarding Adaboost classifier

[Guillaume Lemaître]

I want just to recap a few things:

> I need to train the classifier using adaboost classifier to obtain Haar
features from image dataset
> So can you suggest any method to extract features from image(24*24) datase

You just mentioned what was your requirement regarding the feature to
extract -> Haar features.
My feeling is that you want to reimplement the paper of Viola and Jones for
face detection.

So you could check with the folks of scikit-image if they have something
related -> https://github.com/scikit-image/scikit-image/pull/1444
You could also check opencv which offer functions, classe, and helper ->
http://docs.opencv.org/trunk/d7/d8b/tutorial_py_face_detection.html /
http://docs.opencv.org/2.4/modules/objdetect/doc/cascade_classification.html

At the end, sklearn can help you with the AdaBoostClassifier, ranking of
the features, and the evaluation of the pipeline.


On 19 March 2017 at 07:57, Afzal Ansari <b113053 at iiit-bh.ac.in> wrote:

> Thank you for your response. First I want to extract useful features from
> images so as to get n_features. So can you suggest any method to extract
> features from image(24*24) dataset? Then I can possibly train the
> classifier.
>
> Thanks.
>
> On Sun, Mar 19, 2017 at 11:49 AM, Jacob Schreiber <jmschreiber91 at gmail.com
> > wrote:
>
>> You really need to provide more details with what exactly you're stuck
>> with. If you've extracted useful features from some image into a matrix X
>> with binary labels y you can just do `clf.fit(X, y)` to train the
>> classifier.
>>
>> On Sat, Mar 18, 2017 at 10:21 PM, Afzal Ansari <b113053 at iiit-bh.ac.in>
>> wrote:
>>
>>> Hello Sir,
>>>  I want to classify images containing negative and positive samples
>>> using Adaboost classifier. So how can I do that classification? Please help
>>> me regarding this.
>>>
>>> Thanks.
>>>
>>> On Sat, Mar 18, 2017 at 11:03 PM, Francois Dion <francois.dion at gmail.com
>>> > wrote:
>>>
>>>> You need to provide more details on exactly what you need. I'll take a
>>>> stab at it:
>>>>
>>>> Are you trying to replicate OpenCV cascade training?
>>>> If so, what they call DAB is Scikit learn adaboostclassifier (
>>>> http://scikit-learn.org/stable/modules/generated/sklearn.en
>>>> semble.AdaBoostClassifier.html)‎ with algorithm=SAMME.
>>>> RAB is SAMME.R.
>>>>
>>>>
>>>> ‎Francois
>>>>
>>>>
>>>> Sent from my BlackBerry 10 Darkphone
>>>> *From: *Afzal Ansari
>>>> *Sent: *Saturday, March 18, 2017 00:51
>>>> *To: *scikit-learn at python.org
>>>> *Reply To: *Scikit-learn user and developer mailing list
>>>> *Subject: *[scikit-learn] Regarding Adaboost classifier
>>>>
>>>> Hello Developers!
>>>>  I am currently working on feature extraction method which is based on
>>>> Haar features for image classification. I am unable to find pure
>>>> implementation of adaboost classifier algorithm on the internet even on
>>>> scikit learn web. I need to train the classifier using adaboost classifier
>>>> to obtain Haar features from image dataset.
>>>> Please help me regarding this code. Reply soon.
>>>>
>>>> Thanks in advance.
>>>>
>>>>
>>>> _______________________________________________
>>>> scikit-learn mailing list
>>>> scikit-learn at python.org
>>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>>
>>>>
>>>
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>>
>>
>> _______________________________________________
>> scikit-learn mailing list
>> scikit-learn at python.org
>> https://mail.python.org/mailman/listinfo/scikit-learn
>>
>>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
>


-- 
Guillaume Lemaitre
INRIA Saclay - Parietal team
Center for Data Science Paris-Saclay
https://glemaitre.github.io/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20170319/9e14c426/attachment-0001.html>