File Exchange

image thumbnail

Read RAW camera images

version 21.02 (654 KB) by Emmanuel Farhi
Read RAW camera images from within Matlab, using DCRAW or LibRAW


Updated 03 Feb 2021

From GitHub

View license on GitHub


Read RAW camera images from within Matlab, using DCRAW or LibRAW.

Image of READRAW

The use of this class boils down to simply creating the object. Then, you may simply use the imread and imfinfo call as usual, and RAW files will magically be handled.

In the following example, we just call readraw once, and then all is done with imread and imfinfo as you would do with other image formats.

im = imread('file.RAW'); % this creates a file.tiff
info = imfinfo('file.RAW');
delete('file.tiff'); % to save disk space, if ever needed

The default DCRAW setting for the importation is '-T -4 -t 0 -v' to get the raw data.

NOTE: Each RAW file will be converted to a 16-bits TIFF one at the same location as the initial RAW file. This file is then read again by imread to actually get the image RGB channels. If you have created these files (which are each 146 Mb for 6x4k images), you may either remove them, or further access them without requiring conversion.

Supported RAW camera image formats include:
RAW File Format Description
CRW, CR2 Canon digital camera RAW file formats
NEF Nikon digital camera RAW file format
ORF Olympus digital camera RAW file format
RAF Fuji digital camera RAW file format
RWL Leica camera RAW file format
PEF, PTX Pentax digital camera RAW file format
X3F Sigma digital camera RAW file format
DCR, KDC, DC2, K25 Kodak digital camera RAW file format
SRF, ARW, MRW, MDC Sony/Minolta digital camera RAW file format
RAW Panasonic, Casio, Leica digital camera RAW file format
DNG (CS1, HDR) Adobe RAW file format (Digital Negative)
BAY Casio RAW (Bayer)
ERF Epson digital camera RAW file format
FFF Imacon/Hasselblad RAW format
PXN Fotoman RAW
RDC Ricoh RAW format

If you wish to import the RAW files with specific DCRAW options, use the readraw class method 'imread' with options as 3rd argument e.g:

dc = readraw;
im = imread(dc, 'file.RAW', '-a -T -6 -n 100');

and if you wish to get also the output file name and some more information:

[im, info, output] = imread(dc, 'file.RAW', '-T -4 -t 0 -v');

Some useful DCRAW options are:

-T write a TIFF file, and copy metadata in
-w -T -6 -q 3 use camera white balance, and best interpolation AHD
-a -T -6 use auto white balance
-T -4 use raw data, without color scaling, nor white balance
-i -v print metadata
-z set the generated image date to that of the camera
-n 100 remove noise using wavelets
-w use white balance from camera or auto
-t 0 do not flip the image


readraw class instantiation. No argument.
compile check for DCRAW availability or compile it
delete remove readraw references in imformats
imread read a RAW image using DCRAW. Allow more options
imfinfo read a RAW image metadata using DCRAW


Copy the directory and navigate to it. Then type from the Matlab prompt:


READRAW can use any of the following installed tools:

libraw (with header/include files)
libraw binary tools (dcraw_emu, raw-identify, simple_dcraw)

which should be in the executable search path. These can be installed for Debian-like systems with:

sudo apt install dcraw libraw-bin exiv2 metacam exifprobe libraw-dev
sudo ln -s /usr/lib/libraw/dcraw_emu /usr/local/bin
sudo ln -s /usr/lib/libraw/raw-identify /usr/local/bin

To make use of the libraw direct reader (which is slightly faster than the others), you may have to compile the MeX file for your system by running:

cd path-to-readraw/private

which makes use of the excellent contribution from E. Segre.

If no RAW reader is found, and DCRAW is not yet installed on the computer, you will need a C compiler. The DCRAW C source file (provided with READRAW, private directory) will be built and used.

DCRAW is a great tool
Reading RAW files into MATLAB and Displaying Them
RAW Camera File Reader by Bryan White 2016
LibRaw MeX and

License: (c) E. Farhi, GPL2 (2018)

Cite As

Emmanuel Farhi (2021). Read RAW camera images (, GitHub. Retrieved .

MATLAB Release Compatibility
Created with R2010a
Compatible with any release
Platform Compatibility
Windows macOS Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!
To view or report issues in this GitHub add-on, visit the GitHub Repository.
To view or report issues in this GitHub add-on, visit the GitHub Repository.