块均值哈希方法

Question

在Block Mean Value Based Image Perceptual Hashing中，边杨提出了一种散列方法：

第一个感知哈希函数基于均值

a) Normalize the original image into a preset sizes;

b) Divide the size-normalized image I into non overlapped blocks I1, I2, …, IN, 
in which N is the block number equal to length of the final hash bit string;

c) Calculate the mean value sequence {M1, M2, …, MN} from corresponding 
block sequence {I’1, I’2, …, I’N } and obtain the median value Md of 
this sequence as: Md = median (Mi) (i=1,2,…, N) 

d) Normalize the mean value sequence into a binaryform and obtain the hash 
values h as:
h(i)=0 if Mi<Md
h(i)=1 if Mi>= Md

我的问题是部分“b”，任何人都知道如何将图像分成块？是这样的吗？