Why you perform unsupervised clustering on each feature and how can you calculate the iTED_Features?

[yunfei920406]

For example, if one feature (entropy) is clustered into four clusters (In this case, the k was determined as four). How can you then calculate the iTED_Entropy?

[yunfei920406]

According to your paper, if one hundred features were extracted and the VOI was segemented into 50 subregions. We can obtain a 10050 2D array. How can we then obtain the 1001 array which is iTED_Features.