Jump to content

Data Transformer for high ISO


t024484

Recommended Posts

Advertisement (gone after registration)

It stroke me that the black level value for all ISO values gradually change from top to bottom of the image, to an amount of 4%, instead of being stable over the whole Image.

This causes an error in black level correction, because only one and the same amount is used for correction.

This black level is to be subtracted when conversion takes place in the RAW converter.

At lower ISO values the error caused by this is completely unimportant, because the black level at ISO 160 is 44, so 4% of 44 is only 1.8.

But the higher the ISO value, the higher the black level, being 640 at ISO 2500.

4% of 640 = 25.6 seems already quite an amount being introduced as an error.

So instead of subtracting 640 of all values from top to bottom of the image, subtracting should really start at 640 at the top and gradually change to 599 at the bottom of the image.

 

That is exactly what the program does what I have made available.

It reads the file, proposes the correction values based upon the information within the raw file, and makes the corrections.

 

It generates a new file having the name of the old file with a D added in front, so the old file remains untouched.

To play around, all values can be changed, X being the correction at the top of the Image, and Y at the bottom. In the converted file, the black level to be subtracted is set to zero, to inform the Raw processor that no more correction has to be done.

 

http://dl.getdropbox.com/u/2008358/DataTransformer1.zip

 

It would be nice to get some feedback if this correction makes any difference or not.

The use is restricted to uncompressed M9 files right from the camera, and is targeted at high ISO values although it works for all ISO values.

 

Hans

Link to post
Share on other sites

It stroke me that the black level value for all ISO values gradually change from top to bottom of the image, to an amount of 4%, instead of being stable over the whole Image.

 

Hans, do you have an explanation as to why this is the case?

Link to post
Share on other sites

Hans, do you have an explanation as to why this is the case?

 

I can only guess why this is the case. Either the offset voltage that is added to the sensor signal seems to be "leaking", or the sensor signal is changing because of the temperature increase during readout ???

Unfortunately, there is no specification sheet for this sensor to clarify.

 

Hans

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...