Location:
Search - IDCT2
Search list
Description: to understand the Algorithm go to matlab help in page dct2 and idct2 to get the mathematical expression
for M = N = 8, we can calculate the most of hard values and save it as LUTs to speed up the execution
now compare our special 8X8 functions with the internal general functions
use this code:
A = int32(100*rand(8,8))
tic for i = 1 : 1000 IDCT_8X8(DCT_8X8(A)) end toc
tic for i = 1 : 1000 idct2(dct2(A)) end toc
i had built the general functions too, but with low speed
Platform: |
Size: 1180 |
Author: hujik |
Hits:
Description: to understand the Algorithm go to matlab help in page dct2 and idct2 to get the mathematical expression
for M = N = 8, we can calculate the most of hard values and save it as LUTs to speed up the execution
now compare our special 8X8 functions with the internal general functions
use this code:
A = int32(100*rand(8,8))
tic for i = 1 : 1000 IDCT_8X8(DCT_8X8(A)) end toc
tic for i = 1 : 1000 idct2(dct2(A)) end toc
i had built the general functions too, but with low speed -to understand the Algorithm go to matlab help in page dct2 and idct2 to get the mathematical expressionfor M = N = 8, we can calculate the most of hard values and save it as LUTs to speed up the executionnow compare our special 8X8 functions with the internal general functionsuse this code: A = int32 (100* rand (8,8)) tic for i = 1: 1000 IDCT_8X8 (DCT_8X8 (A)) end toc tic for i = 1: 1000 idct2 (dct2 (A)) end toc i had built the general functions too, but with low speed
Platform: |
Size: 1024 |
Author: hujik |
Hits:
Description: CT2 IDCT2 变换C代码。经调试成功，适用于altera,有结果。-CT2 IDCT2 transform C code. After successful testing for altera, bear fruit.
Platform: |
Size: 1171456 |
Author: 金夕 |
Hits:
Description: M=512 原图像长度
N=64
K=8
I=zeros(M,M)
J=zeros(N,N)
BLOCK=zeros(K,K)
显示原图像
I=imread( f:\ceshi\yuanshitu.bmp )
figure
imshow(I)
xlabel( 原始图像 )
显示水印图像
J=imread( f:\ceshi\logistic_nuaa64.bmp )
figure
imshow(J)
xlabel( 水印图像 )
嵌入水印
tem=1
for p=1:N
for q=1:N
x=(p-1)*K+1
y=(q-1)*K+1
BLOCK=I(x:x+K-1,y:y+K-1)
BLOCK=dct2(BLOCK)
if J(p,q)==0
a=-1
else
a=1
end
BLOCK(2,1)=BLOCK(2,1)*(1+a*0.01)
BLOCK=idct2(BLOCK)
I(x:x+K-1,y:y+K-1)=BLOCK
end
end
显示嵌入水印后的图像
figure
imshow(I)
xlabel( 嵌入水印后的图像 )
imwrite(I, f:\ceshi\embedded.bmp )-M=512 原图像长度
N=64
K=8
I=zeros(M,M)
J=zeros(N,N)
BLOCK=zeros(K,K)
显示原图像
I=imread( f:\ceshi\yuanshitu.bmp )
figure
imshow(I)
xlabel( 原始图像 )
显示水印图像
J=imread( f:\ceshi\logistic_nuaa64.bmp )
figure
imshow(J)
xlabel( 水印图像 )
嵌入水印
tem=1
for p=1:N
for q=1:N
x=(p-1)*K+1
y=(q-1)*K+1
BLOCK=I(x:x+K-1,y:y+K-1)
BLOCK=dct2(BLOCK)
if J(p,q)==0
a=-1
else
a=1
end
BLOCK(2,1)=BLOCK(2,1)*(1+a*0.01)
BLOCK=idct2(BLOCK)
I(x:x+K-1,y:y+K-1)=BLOCK
end
end
显示嵌入水印后的图像
figure
imshow(I)
xlabel( 嵌入水印后的图像 )
imwrite(I, f:\ceshi\embedded.bmp )
Platform: |
Size: 1024 |
Author: 123096321 |
Hits:
Description: u can calculate DCT
clc
inp = imread( Im1.jpg )
inp1 = imread( Im2.jpg )
A = double(inp(:,:,1))
B = double(inp1(:,:,1))
A1=double(blkproc(A,[8 8], dct2 ))
B1=double(blkproc(B,[8 8], dct2 ))
[r,c] = size(A1)
r1 = []
for i=1:r
for j=1:c
if A1(i,j) > B1(i,j)
r1(i,j)=A1(i,j)
else
r1(i,j)=B1(i,j)
end
end
end
r1= double(r1(:,:,1))
C=blkproc(r1,[8 8], idct2 )
round(C)
C1=idct2(r1)
figure
imshow(uint8(C))
figure
imshow(uint8(C1))
imwrite(uint8(C), dct8x8min.jpg ) -u can calculate DCT
clc
inp = imread( Im1.jpg )
inp1 = imread( Im2.jpg )
A = double(inp(:,:,1))
B = double(inp1(:,:,1))
A1=double(blkproc(A,[8 8], dct2 ))
B1=double(blkproc(B,[8 8], dct2 ))
[r,c] = size(A1)
r1 = []
for i=1:r
for j=1:c
if A1(i,j) > B1(i,j)
r1(i,j)=A1(i,j)
else
r1(i,j)=B1(i,j)
end
end
end
r1= double(r1(:,:,1))
C=blkproc(r1,[8 8], idct2 )
round(C)
C1=idct2(r1)
figure
imshow(uint8(C))
figure
imshow(uint8(C1))
imwrite(uint8(C), dct8x8min.jpg )
Platform: |
Size: 29696 |
Author: parthiban |
Hits:
Description: 对frame1进行一个简化的帧内编码方案: A)将图像分为8x8的块. B)将图像每一块做DCT2变换. C)将DCT系数采用11.ppt page 35的亮度量化表量化. 量化系数 = 取整 (DCT系数./量化表). D) 统计整个图像中非零量化系统的比重, 该比重可反映什么? E) 将量化系数进行 IDCT2, 绘制出重建图像.-a example for matlab to deal with a picture
Platform: |
Size: 1024 |
Author: 姚晓栋 |
Hits:
Description: 对“2.1.bmp”图像操作，将其放大至原来大小的1.5倍，采用双线性插值方法放大。
对“2.1.bmp”绕图像中心逆时针旋转20度，采用双线性插值方法。此题目不要使用imresize和imrotate函数，自己写程序完成。
2.对畸变图像进行矫正:”2.2.bmp”为畸变后的图像，已知畸变后图像上的四个点（118，88）（90，140）（139，168）（168，116）对应畸变前的点分别为为（98，98）（98，158）（158，158）（158，68），用对应点约束法求畸变前的图像。（题目中的坐标为行-列坐标；涉及到的插值用双线性插值法）
3.利用离散余弦变换函数（dct2和idct2）对“lena.bmp”做压缩，观察压缩效果。
-Image manipulation, "2.1.bmp" zoom to 1.5 times the original size, using bilinear interpolation method to enlarge.
"2.1.bmp" around the center of the image counterclockwise rotation of 20 degrees, a bilinear interpolation method. This topic Do not use imresize, and imrotate function, write your own program to complete.
2 of the distortion of the image to be corrected: "2.2.bmp" image after distortion, it is known that the four points on the image of the distortion (118,88) (90,140) (139,168) (168,116) corresponding to distortion prior to the point were to (98,98) (98,158) (158,158) (158,68), with a corresponding point constraint method seeking the image in front of the distortion. (Coordinates in the subject line- column coordinate involves interpolation with bilinear interpolation)
3 using discrete cosine transform the function (dct2, and idct2) "lena.bmp" do the compression, to observe the compression effect.
Platform: |
Size: 2048 |
Author: 单车少年 |
Hits:
Description: 利用离散余弦变换函数（dct2和idct2）模拟JPEG压缩过程。 对“lena.bmp”做压缩，观察压缩效果。（JPEG压缩过程： 首先把原始的图像划分为8×8的子图像，然后利用dct2分别
对各个子图像进行离散余弦变换，对变换系数仅保留左上角15个值， 然后利用这15个系数进行离散余弦反变换重新得到各个子图像，再 将子图像拼接成完整图像） -Using discrete cosine transform function (dct2 and idct2) analog JPEG compression process. On " lena.bmp" do compression, the compression effect was observed. (JPEG compression process: First, the original image is divided into 88 sub-image, and then use dct2 of each sub-image, respectively, discrete cosine transform, the upper left corner of the transform coefficients to retain only 15 values, then use the 15 coefficients discrete cosine transform regain each sub-image, and then sub-images together into a complete picture)
Platform: |
Size: 1024 |
Author: 付兴银 |
Hits:
Description: 用解析式定义或矩阵式定义自行编写进行二维DCT和离散沃尔什-哈达玛变换的函数。
在Matlab中对给定图像进行二维离散傅立叶变换，并显示其居中的频谱图，可用Matlab自带函数进行实验。
分别用Matlab自带的dct2()和idct2()对给定图像进行二维DCT变换，并以图像的形式显示其变换系数矩阵。
-And the preparation of two-dimensional DCT is defined analytically discrete Walsh matrix or a self-defined- Hadamard transform function. For a given two-dimensional discrete Fourier transform in Matlab image, and displays its spectrum centered available Matlab function comes experiment. respectively two-dimensional DCT transform for a given image DCT2 comes with Matlab () and idct2 (), and displays an image thereon in the form of matrix transform coefficients.
Platform: |
Size: 166912 |
Author: 吕文 |
Hits: