How to perform classification

When it comes to Classification of 4 different sets images,what methods should be applied? I mean after calculating all the features of the four sets images and storing them into a database....How to differentiate them using classification?(I have the target data (feature vector) already calculated).Plz guide me.

Answer 1

[the cyclist]

I suggest you read this documentation about classification in MATLAB.

[Durga]

Thank U Sir....One more query on this part Sir, In case of a 2-class classification, after classification,the output is either 0 or 1, Whereas zero signifies one set and 1 signifies another set.

But when it comes to identification of objects more than 2,then which method can be applied and how the output will be generated(i.e,whether theoutput will be generated in 0 or 1 or something else to signify the detected object(class).

[Image Analyst]

The example on the page dealt with 3 classes. Generally classification routines give you a class number like 0,1,2,3,4,... or 1,2,3,4,5... When classifying images, you'll get an image out where the value of the pixel is the value of the class it assigned. For some methods, like kmeans() you need to create the image yourself. For you, is one whole image one class (like you have photos of either sky or grass)? Or can it have multiple classes in the image (like your images have both sky and grass in each image)?

[Durga]

Sir,I have the whole image as a class(Oral Disease Images,4 different Oral Disease Image sets are there)...

and i have calculated features of the * disease effected area * of each image..now i need to classify those and detect each set correctly..