fruits recognition with KNN uisng color features

Dear Sir,

I have extracted features for fruits recognition system using fruits_360 dataset.

color features i selected are

1.hsv with statistical measurement

2.yCbCr with statistical measurement

3.Lab with statistical measurement

4.CCV

and finally wrote all the feature vectors in excel. i have taken 15 category of fruits image with 20 samples in each category. i have used 60% and 40% ratio for classification. THen performing the classification with KNN. but i got less accuracy.

I need to know the following things

1. the codewriiten was correct or not?
2. if its correct means what may be reason for poor accuracy
3. if its not correct maens please tell me the way to solve it

please gothrough the codes and help me