Image Compression: How to use blockproc function with dct2 inside

Question

Fady Yousif 2022-3-11

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1669064-image-compression-how-to-use-blockproc-function-with-dct2-inside

回答： Abolfazl Chaman Motlagh 2022-3-11

For the uploaded image below with 536x704 pixel size, my code reads the image, converts to ycbcr and subsamples, then with dct tries to make 8x8 block with 4:2;0 format and choose the top 6th row of the luminance.

My code:

I = imread('Flooded_house.jpg','jpg');

%converting to ycbcr

YCbCr = rgb2ycbcr(I);

Y = YCbCr(:,:,1);

Cb = YCbCr(:,:,2);

Cr = YCbCr(:,:,3);

%sub-sampling

Sub_Cb = Cb(1:2:end, 1:2:end);

Sub_Cr = Cr(1:2:end, 1:2:end);

%Now computing the 8x8 block DCT transform coefficients in 4:2:0 format

DCT_Trans = @(block_struct) dct2(block_struct.data);

DCT_Y = blockproc(Y, [8 8], DCT_Trans);

DCT_Cb = blockproc(Sub_Cb, [8 8], DCT_Trans, 'PadPartialBlocks', true);

DCT_Cr = blockproc(Sub_Cr, [8 8], DCT_Trans, 'PadPartialBlocks', true);

%Part (a)

%putting the first block of the DCT image

First_Block = DCT_Trans(1:8,1:8);

imshow(First_Block)

%computing the first 2 blocks in the 6th row from the top of the luminance

%component

Sixth_Block = DCT_Trans(41:48,1:8);

imshow(Sixth_Block)

Sixth_Block_p2 = DCT_Trans(41:48,9:16);

imshow(Sixth_Block_p2)

Error Code:

Error using Encoder_a>@(block_struct)dct2(block_struct.data)

Too many input arguments.

Error in Encoder_a (line 31)

First_Block = DCT_Trans(1:8,1:8);

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Abolfazl Chaman Motlagh 2022-3-11

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1669064-image-compression-how-to-use-blockproc-function-with-dct2-inside#answer_915044

在 MATLAB Online 中打开

you created DCT_Trans function for blockprop . it's a function which takes structure as input. if you pass structure S to it's input, it takes dct of S.data. so the line

First_Block = DCT_Trans(1:8,1:8);

and all lines like this, return error. because (1:8,1:8) are two input. and they are not structure.

if you wanna call your transformed image you should call DCT_Y, DCT_Cb and DCT_Cr .

don't you mean :

First_Block = DCT_Y(1:8,1:8);

???