Quantize dct cofficents in matlab - matlab

Question

hello i need to perform quantization to dct cofficents for an image, for a block
size of 8*8 pixles in matlab. can you help me with the syntax, thank you.

Answer 1

There is a built-in function in MATLAB for DCT.
You need the signal processing tool box. Type 'ver' (without quotes) in the MATLAB command to see if you have it.
The code:
image = image; % define your image
[m,n] = size(image); % get size of your image
imvector = reshape(image, m*n, 1); % reshape your image to a vector to compute DCT
imdct = dct(imvector); % compute DCT
imagedct = reshape(imdct,m,n); \ reshape result back to original form of your image

Answer 2

There is an example in the help file as well which is very nice:
I = imread('cameraman.tif');
I = im2double(I);
T = dctmtx(8);
dct = #(block_struct) T * block_struct.data * T';
B = blockproc(I,[8 8],dct);
mask = [1 1 1 1 0 0 0 0
1 1 1 0 0 0 0 0
1 1 0 0 0 0 0 0
1 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0];
B2 = blockproc(B,[8 8],#(block_struct) mask .* block_struct.data);
invdct = #(block_struct) T' * block_struct.data * T;
I2 = blockproc(B2,[8 8],invdct);
imshow(I), figure, imshow(I2)

Answer 3

To quantize DCT coefficients, you simply divide each coefficient by a quantization term and round to integers. The quantization terms are often unique for each coefficient, and are stored in a quantization matrix.
Wikipedia has a nice example. Here is how to implement that example in Matlab.
coef = [
-415 -33 -58 35 58 -51 -15 -12;
5 -34 49 18 27 1 -5 3;
-46 14 80 -35 -50 19 7 -18;
-53 21 34 -20 2 34 36 12;
9 -2 9 -5 -32 -15 45 37;
-8 15 -16 7 -8 11 4 7;
19 -28 -2 -26 -2 7 -44 -21;
18 25 -12 -44 35 48 -37 -3
];
quant = [
16 11 10 16 24 40 51 61;
12 12 14 19 26 58 60 55;
14 13 16 24 40 57 69 56;
14 17 22 29 51 87 80 62;
18 22 37 56 68 109 103 77;
24 35 55 64 81 104 113 92;
49 64 78 87 103 121 120 101;
72 92 95 98 112 100 103 99
];
quantCoef = round(coef ./ quant)
quantCoef =
-26 -3 -6 2 2 -1 0 0
0 -3 4 1 1 0 0 0
-3 1 5 -1 -1 0 0 0
-4 1 2 -1 0 0 0 0
1 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0

Answer 4

clc
clear all
close all
image=imread('Water lilies.jpg');
Q=[8 36 36 36 39 45 52 65;
36 36 36 37 41 47 56 68;
36 36 38 42 47 54 64 78;
36 37 42 50 59 69 81 98;
39 41 47 54 73 89 108 130;
45 47 54 69 89 115 144 178;
53 56 64 81 108 144 190 243;
65 68 78 98 130 178 243 255];
GrayImage=rgb2gray(image);
NewImage=uint8(zeros(size(GrayImage)));
Q=uint32(Q);
for i=1:size(GrayImage,1)/8
for j=1:size(GrayImage,2)/8
block=GrayImage(((i-1)*8)+1:((i-1)*8)+8,((j-1)*8)+1:((j-1)*8)+8);
dct=dct2(block);
dct=uint32(dct);
a=dct./Q;
z=a.*Q;
idct=idct2(z);
NewImage(((i-1)*8)+1:((i-1)*8)+8,((j-1)*8)+1:((j-1)*8)+8)=uint8(idct);
end
end
imwrite(NewImage,'NewImage.jpg');
GrayImage=double(GrayImage);
NewImage=double(NewImage);
MSE=0;
for i=1:size(GrayImage,1)
for j=1:size(GrayImage,2)
d=(GrayImage(i,j)-NewImage(i,j))^2;
MSE=d+MSE;
end
end