num_class+1?

Hi, 
Firstly, thank you for the code. I'm very new to this image segmentation problem and I just want to know what is the reason for having "num_class+1 " in below line.
 annotations_ohe = tf.one_hot(annotations, num_classes+1, axis=-1)
Also, how did you calculate class and mean pixel values?
Please explain a bit briefly. 
Thanks again.