2

I need to train SVM with 2 different features..the problem is one feature is the HOG with length 144 and other an RGB value with length 3.. Can i combine these two features to train SVM and test using the same? the code iam using is

X1=xlsread('train1.xlsx'); Y1=xlsread('CLASS1.xlsx'); svmStruct1=svmtrain(X1,Y1); classes=svmclassify(svmStruct1,test);

the test data contains both RGB and HOG values..

user67593
  • 23
  • 2

1 Answers1

0

Yes, you can do it. Concatenate the two feature vectors to get a single feature vector of length 147; the first 144 components are from the HOG, and the last 3 components are from the RGB. Then apply the SVM on the feature vector of length 147. SVM doesn't care where the individual numbers came from.

Code is off-topic on this site.

D.W.
  • 167,959
  • 22
  • 232
  • 500