Simple DCT problem

Hi all,

I am new to DCT and now testing my FDCT and IDCT program.
I found some of my high frequency DCT part is higher than low frequency
DCT part.
And I don't know whether it is correct.

Where can I find example of 8*8 DCT input and output to test my
code,thanks!

BTW,
my code's DCT input:
55  60  58  55  60  58  55  60
58  55  60  58  55  60  58  55
60  58  55  60  58  55  60  58
55  60  58  55  60  58  55  60
58  55  60  58  55  60  58  55
60  58  55  60  58  55  60  58
55  60  58  55  60  58  55  60
58  55  60  58  55  60  58  55
my code's DCT output:
461  0  -1  0  -1  -1  1  0
0  0  0  0  0  -2  0  0
-1  0  -1  0  -2  -1  2  0
0  0  0  0  0  -2  -1  0
-1  0  -2  0  -3  -3  4  0
1  -2  1  -2  3  -11  -3  1
1  0  2  1  4  3  -5  0
0  0  0  0  0  1  0  0

Any suggestions will be appreciated! 
All the best,
Davy

Davy wrote:

> Hi all,
> 
> I am new to DCT and now testing my FDCT and IDCT program.
> I found some of my high frequency DCT part is higher than low frequency
> DCT part.
> And I don't know whether it is correct.
> 
> Where can I find example of 8*8 DCT input and output to test my
> code,thanks!
> 
> BTW,
> my code's DCT input:
> 55  60  58  55  60  58  55  60
> 58  55  60  58  55  60  58  55
> 60  58  55  60  58  55  60  58
> 55  60  58  55  60  58  55  60
> 58  55  60  58  55  60  58  55
> 60  58  55  60  58  55  60  58
> 55  60  58  55  60  58  55  60
> 58  55  60  58  55  60  58  55
> my code's DCT output:
> 461  0  -1  0  -1  -1  1  0
> 0  0  0  0  0  -2  0  0
> -1  0  -1  0  -2  -1  2  0
> 0  0  0  0  0  -2  -1  0
> -1  0  -2  0  -3  -3  4  0
> 1  -2  1  -2  3  -11  -3  1
> 1  0  2  1  4  3  -5  0
> 0  0  0  0  0  1  0  0
> 
> Any suggestions will be appreciated! 

And what does the inverse of the DCT give from the above?

> All the best,
> Davy

Test it initially with simple patterns that you can work out by hand. 
Walking ones will give you a fairly easy test pattern for DCT that 
should quickly show if your implementation is OK.

1 0 0 0 0 0 0 0
etc.

Regards,
Martin Brown

> > Where can I find example of 8*8 DCT input and output to test my
> > code,thanks!

Simplest way is to use the inbuilt dct&idct functions of MATLAB to
obtain a test vector. 

Good Luck.

Parthasarathy

Hi all,

I use Matlab's dct(dct(x')') and test my code. All OK!

All the best,
Davy

to test use the source code proposed by 
http://skal.planet-d.net/coding/dct.html

note :
- there is lot's of different implementation of the DCT.
- some are very fast but less accurate
- if you deal with an mpeg video std implementation only idct is 
describe and should respect I3E-1394 precision. DCT is the encoding part 
and it's not normalized !!!
- you should find on the web an IEEE-1394 idct tester which test if your 
implentation respect the standard requirement.

good luck

Davy wrote:
> Hi all,
> 
> I am new to DCT and now testing my FDCT and IDCT program.
> I found some of my high frequency DCT part is higher than low frequency
> DCT part.
> And I don't know whether it is correct.
> 
> Where can I find example of 8*8 DCT input and output to test my
> code,thanks!
> 
> BTW,
> my code's DCT input:
> 55  60  58  55  60  58  55  60
> 58  55  60  58  55  60  58  55
> 60  58  55  60  58  55  60  58
> 55  60  58  55  60  58  55  60
> 58  55  60  58  55  60  58  55
> 60  58  55  60  58  55  60  58
> 55  60  58  55  60  58  55  60
> 58  55  60  58  55  60  58  55
> my code's DCT output:
> 461  0  -1  0  -1  -1  1  0
> 0  0  0  0  0  -2  0  0
> -1  0  -1  0  -2  -1  2  0
> 0  0  0  0  0  -2  -1  0
> -1  0  -2  0  -3  -3  4  0
> 1  -2  1  -2  3  -11  -3  1
> 1  0  2  1  4  3  -5  0
> 0  0  0  0  0  1  0  0
> 
> Any suggestions will be appreciated! 
> All the best,
> Davy
> 

Hi Bernie,

Thank you :-)

I use Chen's algorithm (takes 16 Mul). It is 8 times faster than MPEG's
float algorithm on my P3 machine.

Best regards,
Davy