On Jun 11, 6:12 pm, aruzinsky <aruzin... at (no spam) general-cathexis.com> wrote:
On Jun 9, 5:38 am, AJ <arandal... at (no spam) gmail.com> wrote:
Hi All,
I have an image which is very blocky (image can be seen
athttp://img222.imageshack.us/my.php?image=degradedimagesh2.png) and
would like to try and improve it's quality as much as possible.
I understand that the image is of such a poor quality that I do not
expect any huge improvements - that is not my aim, my aim is to see if
it can be improved at all, and if so, which method(s) could yield the
best results.
So far I have tried some iterative restoration techniques such as
Richardson-Lucy deconvolution and Blind deconvolution - both have
proved to give a slight improvement.
I was hoping someone would have some other suggestions as to how I
could possibly improve the image.
Just in case understanding how the image was degraded in the first
place might help lead to a good suggestion for improvement algorithms,
I created the image using the following steps:
* 8x8 Sub-Block and DCT image
* store top 2 coefficients from each sub-block
* use these to create a new image
Many thanks,
AJ
Okay, this improved result,
http://www.general-cathexis.com/images/degradedimagesh2_4.jpg
, was obtained by incorporating some information from the cosine half
wave within each 8 pixel interval. The values are all of the same
sign within each half interval so I can get a fairly natural looking
image by reducing to 128 x 64 using a box kernel. Whereas, a 64 x 64
reduction averages the half cosine to zero, the 128 x 64 reduction
sort of makes it a half square wave.
Here is the reduced image in 16 bit format,
http://www.general-cathexis.com/images/degradedimagesh2_Reduced.png
, which I enlarged 8X using a proprietary nonlinear method and then
reduced the width 0.5X using box interpolation.
Just a quick thank you to everyone who has made the effort to
experiment with my degraded image, I really appreciate it! :)
I have measured the PSNR of each submitted (including some suggestions
from the 'comp.compression' group, which can be found here - 'http://
groups.google.com/group/comp.compression/browse_thread/thread/
7c7b9058de482155#'), against the original copy of the image. The
results are as follows:
Aruzinsky (http://www.general-cathexis.com/images/
degradedimagesh2_2.png) - 25.81 dB
Aruzinsky (http://www.general-cathexis.com/images/
degradedimagesh2_3.png) - 25.51 dB
Aruzinsky (http://www.general-cathexis.com/images/
degradedimagesh2_4.jpg) - 27.91 dB
pisz_na (http://i287.photobucket.com/albums/ll123/ememek/r23.png) -
19.27 dB
jacko (http://indi.hpsdr.com/degradedmagesh11.png) - 20.05 dB
jacko (http://indi.hpsdr.com/image2.png) - 19.37 dB
jacko (http://indi.hpsdr.com/diffusiontop.png) - 22.73 dB
mine (AJ) (http://i287.photobucket.com/albums/ll146/arandall85/
interpolated_iterated_image.png) 24.55 dB
If there was a prize, you would get it Aruzinsky! The only issue I
have got is that ideally I would like to implement the technique
described in Matlab, since the rest of my image processing techniques
are written using it.
Does anyone have any idea how I could go about converting Arunzinsky's
method (shown below) into a format Matlab would understand?:
[snip]"was obtained by incorporating some information from the cosine
half
wave within each 8 pixel interval. The values are all of the same
sign within each half interval so I can get a fairly natural looking
image by reducing to 128 x 64 using a box kernel. Whereas, a 64 x 64
reduction averages the half cosine to zero, the 128 x 64 reduction
sort of makes it a half square wave."[snip]
Many thanks,
AJ- Hide quoted text -
- Show quoted text -
