zeros为创建一个值为零的数组；

如matrix1=zeros(4,5);%4*5的矩阵，矩阵中每个元素都为0

matrix2=zeros(4,5,3);%4*5*3的数组，数组中每个元素都为0

下面举一个将图像存到数组的例子

对RGB图片1.jpg，2.jpg；大小为700*500*3

创建4D空数组700*500*3*2

img=zeros(700,500,3,2);

直接将图片赋予数组会发现不是原图，代码及运行结果如下：

img1=imread(strcat('C:\Users\Administrator\Desktop\','1.jpg'));

img2=imread(strcat('C:\Users\Administrator\Desktop\','2.jpg'));

% img1_1=im2double(img1);

% img2_2=im2double(img2);

img=zeros(700,500,3,4);

img(:,:,:,1)=img1;

img(:,:,:,2)=img2;

% img(:,:,:,3)=img1_1;

% img(:,:,:,4)=img2_2;

figure(1)

subplot(4,2,1),imshow(img1);

subplot(4,2,2),imshow(img2);

% subplot(4,2,3),imshow(img1_1);

% subplot(4,2,4),imshow(img2_2);

subplot(4,2,5),imshow(img(:,:,:,1));

subplot(4,2,6),imshow(img(:,:,:,2));

% subplot(4,2,7),imshow(img(:,:,:,3));

% subplot(4,2,8),imshow(img(:,:,:,4));

运行结果：

如果将图片转为双精度型，则可正常，代码结果如下：

img1=imread(strcat('C:\Users\Administrator\Desktop\','1.jpg'));

img2=imread(strcat('C:\Users\Administrator\Desktop\','2.jpg'));

img1_1=im2double(img1);

img2_2=im2double(img2);

img=zeros(700,500,3,4);

img(:,:,:,1)=img1;

img(:,:,:,2)=img2;

img(:,:,:,3)=img1_1;

img(:,:,:,4)=img2_2;

figure(1)

subplot(4,2,1),imshow(img1);

subplot(4,2,2),imshow(img2);

subplot(4,2,3),imshow(img1_1);

subplot(4,2,4),imshow(img2_2);

subplot(4,2,5),imshow(img(:,:,:,1));

subplot(4,2,6),imshow(img(:,:,:,2));

subplot(4,2,7),imshow(img(:,:,:,3));

subplot(4,2,8),imshow(img(:,:,:,4));

结果