Classification Image using K Nearest Neighbours

[EDIT: Thu May 12 22:14:37 UTC 2011 Duplicate Removed - MKF]

hello , I'm trying to classify data from a picture using nearest neighbor method Class = knnclassify(Sample, Training, Group) problem is how i get Group

10x .

[Andrew Newell]

This question doesn't make any sense. Could you please provide the relevant part of your code and describe in more detail what your question is?

[Walter Roberson]

I believe the question is how to find the optimal number of clusters for KNN.

[ops OPS]

okay Walter , exemple k means http://www.mathworks.com/matlabcentral/fileexchange/12979-clusterimg

i want it with KNN .i hope you understand me .

[ops OPS]

clustering on an input RGB (m x n x 3) using K Nearest Neighbours .

Answer 1

[xiaolong]

i just have the similar question like you ? i have segmented the image into different region ?but i don't know how to classification the region to different class ?who can help me ?

[xiaolong]

i don't know why the answer is good ？

[Fadi Alsuhimat]

Me to, I don't why this answer is good????

[Image Analyst]

It's not an answer. I unaccepted it. I have used knnsearch(). I'm attaching a simple demo for (x,y) data. If you need a demo, post your image and explanation in a new question.