IA does read it correctly but i reports as being :

102.png [1:1] Real part , linear representation ....

Almost all menus are ghosted ...
incluing the "crop" one ...
One of the active menus available is the "Pixel format conversion -> Conver to Grayscale"
But doing this I will loose half of the data ( I think) ...

Is this the expected behaviour for the PNG 16 bits Gray ?

This gray image is the result of a "dcraw" image conversion starting from a PPM file and converted to a PNG ...
Nonetheless I still have not figured out why some machines do read PPM files and others dont ...

I appreciate the straight forward simplicity of the TWAIN scanning functionality. However still a few remark and some suggestions.

When opening the dialog to scan from TWAIN source and then changing mind (cancel, or cancel after preview), the program gives an error "Failed to access data source". Since i cancelled the operation myself, no error should appear.
Program version: 1.32
Operating system: windows 7 profesional, 64bit
(Version 1.31 even used to hang on this)

Few small suggestions as well:
I believe that it would be better to add the "advanced" scanning options to the main dialog: they don't take much place, the dpi is pretty basic too and the "source type" in advanced settings is duplicate with the choice in the main dialog.
I'd also appreciate if my choice of dpi would be remembered or i could set default choice in some way.

Pieter

Others machines simply report with an "unsupported pixel format"
or with a strange dialog defining a rasterized bmp by selecting the number of pixels vs bit depth ...

In order  to test this I did tried the same dcraw, same image , same Image analyser version ...

I avoid all of those problems by forcing the TIFF flag in dcraw ...
Any idea why the different behaviour ?

Now a small request :
Would be possible to make a warning ( at opening or different info in the mdi window) when reading 16bits TIFF or 16bits PPM  instead of displaying the filename and "24 bit" ....
I know the image is always showing in 24 bit but sometimes I do have several images (some JPEGs others in TIFF or PPM) with the same picture ...
 
Now for something different:
I visually noticed differences between images (specially low lightining conditions) treated from 16 TIFF files  and 8bits JPEGs inside Image Analyser...
So IA still rocks

Horizontal row
input red,
input green,
input blue

Vertical column
output red
output green
output blue

Think of the (basic 3x3) color-matrix as a linear transformation (gain or attenuation) between r,g,b channels.
So each entry is the gain if greater than 1 or attenuation if less than 1 between input and output channels.

r->r, g->r, b->r , r->g, g->g, b->g , r->b, g->b, b->b

But note that is possible to mix channels for example to change green to red if you want by using the corresponding matrix entry.

Or yellow to red (by using the g->r and/or r->g possibly with negative values).

An example is of this is something like in

http://www.adobe.com/devnet/flash/artic … ns_04.html

Would it be possible to implement such a way to edit the colors ?

It is just an new idea ...

I'm interested in Statistics map transformation.
(Menu Operations -> Transformations -> Statistics map..)

I apply this function to PCB(printed cirtuit board) image, I got the image which suitable for mask pattern in shape analysis.

What is the statistics map transformation operation performing ?

I was search internet, but not found for Statistics map transformation.

Thanks for any information about this, if you are able to supply a reference to a paper on this operation i would be grateful.

Although the image is visually correct,  but what is happening ?
Is IA interpreting 8bits from each channel ?
is it averaging the data ?
or simply discarding the remaining PPM data ?

Nonetheless this is still the best

I know I did ask this before but what about 48bits support ?

http://i301.photobucket.com/albums/nn61 … ifacts.jpg

below is the original picture state (shown resampled to 1000 x750)

http://i301.photobucket.com/albums/nn61 … ifacts.jpg

The artifacts make it difficult to know what is really happening to these type of images when editing. Any help appreciated.

I run Image Analyzer on a 14" lcd, hp laptop running vista.

The pics are from a Panasonic Lumix fz35.

Some of the times I have to go back and forth between
"color balance options" and "color Mapper"...
Most of the time I use the "color mapper" histogram grid to help me making the correct color 
adjustments...

In the most recent versions I can see both "histogram and curves" adjustments in
"color balance" option but they are too awkward to be usable ...

Would it be possible to "colorify" the histogram in "color mapper" option ?

I also can see some "method"s but I have no idea of the implications ...
I have "learn" to use some of them but others are still an incognito to me ...

Nonetheless some help (with some tooltips ) would be nice regarding the different "methods"
available in "color balance".

Resuming :
would be possible to select and edit other colorspaces besides R G B in the "color
mapper" interface ?

Of course I do not know if some of the other operations or selections like "gamma" are applyable in some of the the other colorspaces ...

Another issue :
I know that I already asked about this in the past :
would Image analyser in the future be able to read 48 bit images and work in 16 bits per
channel ?

Thanks for the pacience and the best freeware image program

THanks in advace