Classification using KNN

[CODE title="Classification using KNN"]clc;clear;close all;warning off all;



load HOG_features



% imaqreset;



% vid = videoinput('winvideo', 1, 'YUY2_640x480');

% src = getselectedsource(vid);

% vid.FramesPerTrigger = Inf;

% vid.ReturnedColorspace = 'rgb';

% start(vid);

% preview(vid);



%% Get snapshot and process frame image



% h = msgbox('Show Sign !');

% uiwait(h);

%

% im = getsnapshot(vid);



[f,p] = uigetfile('*.*');

im = imread([p,f]);



imshow(im);



% stop(vid);



cform = makecform('srgb2lab');

J = applycform(im,cform);

K=J,:,2);

L=graythresh(J,:,2));

BW1=im2bw(J,:,2),L);

BW1=bwareaopen(BW1,500);

BB=regionprops(BW1,'Boundingbox');

pause(0.01)



gr = rgb2gray(im);



object = imcrop(gr,BB(1).BoundingBox);

object = imresize(object,[100,50]);

figure(2);imshow(object);



[feat] = hog_feature_vector(object);



class = knnclassify(feat,data,group);



switch class

case 0

tts('Sign of 0');

case 1

tts('Sign of 1');

case 2

tts('Sign of 2');

case 3

tts('Sign of 3');

case 4

tts('Sign of 4');

case 5

tts('Sign of 5');

case 6

tts('Sign of 6');

case 7

tts('Sign of 7');

case 8

tts('Sign of 8');

case 9

tts('Sign of 9');



end



imaqreset[/CODE]

İş görür saol

helal arap iş görmez bu

eyvallah işe yarar

işime yarar eyw

yeyx

İş görür saol

ne işe yarar bu

pca mı?

Yararlı paylaşım teşekkürler

Bom fórum muito util.

ne işe yarıyor tam olarak

KNN algorithm is used to classify by finding the K nearest matches in training data and then using the label of closest matches to predict. Traditionally, distance such as euclidean is used to find the closest match

İş görür saol

Ne için bu?