6by9 / dcraw Goto Github PK

LibRaw/LibRaw#401

Latest comment on an issue started in LibRaw (see link above) has requested help in leveraging whatever information about Raspberry Pi camera (HQ 12Mp) raw already done here by done 6by9 into their coding. Anyone here up to the task in assistance? It would be a big step in getting this camera's raw format recognizable with more mainstream applications.

dear 6by9 thank you for sharing your work!

I use the RP HQ Cam (IMX477):

libcamera-raw -t 1000 --ev -10 --awbgains 1.0,1.0 --gain 1.0 --analoggain 1.0 --denoise off --rawfull 1 --shutter 9000 --mode 4056:3040:12:U -o /mnt/rd/test%05d.raw --segment 1

I would like to extract only radiometric information of the camera sensor .raw file.
Basically I want a matrix which exactly shows me the irradiance intensitity of every photocell/pixel.

Which command would you suggest me according to my use case?

I tried something like:

./dcraw -D -6 test0.raw

Cannot decode file test0.raw (https://ufile.io/qh74d895)

I dumped the jpeg+raw image from RPi v1 camera. After I process the images using
dcraw -v -4 -H 0 -W -w -q 3 -T -p embed -o sRGB myimage.jpeg, I obtain a 16bit tiff image with the maximum value being 2^16. I don't understand how and where does the conversion from 10bit to 16bit image takes place inside the code. Can you kindly let me know how to dump demosaiced images that have maximum values around 2^10?

Hi, me again!

I'm really a noob on linux system. Please bear with me.
in the example you gave on doubling 640x200 to 640x400, it was just one single image. How do i do it in batch? like a few hundreds of .ppm to be doubled all at once.

Thank you!

in lines 1188/1190 there are two Huffmann tree tables.
They have the same lengths but different leaf-values.
The first has 13 leaf values, the secand has 14 leaf values.(The latter corresponds to my manual construction of the Huffmann tree)
Two trees with the same lengths must have the same number of leaves!
Doesn't the make_decode read a value beyond the last element in the array? Does the value, it finds there, make any sense?
So, what is the correct missing last value of the first tree?

sudo apt-get install libjasper-dev libjpeg8-dev gettext liblcms2-dev on Bullseye returns:

Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Package libjpeg8-dev is not available, but is referred to by another package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source
However the following packages replace it:
  libjpeg9-dev libjpeg62-turbo-dev

E: Package 'libjpeg8-dev' has no installation candidate

Installing libjpeg9-dev via:

sudo apt-get install libjasper-dev libjpeg9-dev gettext liblcms2-dev

And then running ./buildme produces:

/usr/bin/ld: warning: libjpeg.so.62, needed by /usr/lib/gcc/arm-linux-gnueabihf/10/../../../arm-linux-gnueabihf/libjasper.so, may conflict with libjpeg.so.9

I purged libjpeg9-dev and installed libjpeg62-turbo-dev and the warning didn't pop up and the build seemed to succeed.

Hi, the modified version of dcraw.c does not seem to be installed on my machine.

I've tried several times.
it always gives me the following warning:

warning: libjpeg.so.62, needed by /usr/lib/gcc/arm-linux-gnueabihf/6/../../../arm-linux-gnueabihf/libjasper.so, may conflict with libjpeg.so.8

tried to use dcraw after done with ./buildme, it says command not found

dcraw.c:77:10: fatal error: jasper/jsaper.h: No such file or directory

compilation terminated