Scaling (Normalise) in Support Vector Machine Image Classification

Hi all;

I read Hsu et al. (2003) 'A Practical Guide to Support Vector Classification' and they proposed procedures in SVM. One of them is conduct simple scaling on the data before applying SVM. The purpose is to avoid attributes in greater numeric ranges dominating those in smaller numeric ranges. I have a question, do the implementation of SVM in Matlab using fitcsvm and fitcecoc already contain scaling for the dataset (ex:for image classification) or we need to do that before running the fitcecoc function? Thank you in advance.