This repository contains a collection of OpenCL kernels for image processing. The CLIJ is build on top of it allowing ImageJ / Fiji users in doing GPU-accelerated image processing without the need for learning OpenCL.
If you use it, please cite it:
Robert Haase, Loic Alain Royer, Peter Steinbach, Deborah Schmidt, Alexandr Dibrov, Uwe Schmidt, Martin Weigert, Nicola Maghelli, Pavel Tomancak, Florian Jug, Eugene W Myers. CLIJ: GPU-accelerated image processing for everyone. Nat Methods (2019) doi:10.1038/s41592-019-0650-1
OpenCL offers several pixel types, such as uint8, unit16 and float.
Theoretically, one has to write OpenCL-kernels specifically for given input- and output-images, such as a kernel for
adding images of type float resulting in a float image and a kernel for adding image of type uint8 resulting in
an image of type float. Furthermore, OpenCL defines images and buffers. However, as both are arrays of pixel
intensities in memory, we wanted to access them in a unified way. As this would result in a ridiculous large number of individual kernel implementations, we used
a dialect where placeholders such as IMAGE_src_PIXEL_TYPE represent the pixel type of the output image.
The following list of placeholders are used at the moment. The name imagename must contain src or dst to differentiate readonly and writeonly images.
| Place holder | Replacement during runtime | |
CONVERT_imagename_PIXEL_TYPE |
clij_convert_char_sat clij_convert_uchar_sat clij_convert_short_sat clij_convert_ushort_sat clij_convert_int_sat clij_convert_uint_sat clij_convert_float_sat |
Convert any number to a given type. |
IMAGE_imagename_TYPE |
__read_only image3d_t __read_only image2d_t __write_only image2d_t __write_only image3d_t __global char* __global uchar* __global short* __global ushort* __global float* |
Two dimensional input image type definition. |
IMAGE_imagename_PIXEL_TYPE |
char uchar short ushort float |
Pixel type definition |
GET_IMAGE_DEPTH(imagename) |
constant number | Image size in Z |
GET_IMAGE_HEIGHT(imagename) |
constant number | Image size in Y |
GET_IMAGE_WIDTH(imagename) |
constant number | Image size in X |
POS_imagename_TYPE |
int2 int4 |
Type of coordinate |
POS_imagename_INSTANCE(pos0,pos1,pos2,pos3) |
int2(pos0, pos1) int4(pos0,pos1,pos2,pos3) |
instantiate variable of coordinate |
READ_imagename_IMAGE |
read_imageui (2d) read_imagef (2d) read_buffer2dc read_buffer2duc read_buffer2di read_buffer2dui read_buffer2df read_imageui (3d) read_imagef (3d) read_buffer3dc read_buffer3duc read_buffer3di read_buffer3dui read_buffer3df |
Read pixel intensity from a given position |
WRITE_imagename_IMAGE |
write_imageui (2d) write_imagef (2d) write_buffer2dc write_buffer2duc write_buffer2di write_buffer2dui write_buffer2df write_imageui (3d) write_imagef (3d) write_buffer3dc write_buffer3duc write_buffer3di write_buffer3dui write_buffer3df |
Write pixel intensity to a given position |
- Image dimensionality is limited to three dimensions.
We acknowledge support by the Deutsche Forschungsgemeinschaft under Germany’s Excellence Strategy (EXC2068) Cluster of Excellence Physics of Life of TU Dresden. This project has been made possible in part by grant number 2021-237734 (GPU-accelerating Fiji and friends using distributed CLIJ, NEUBIAS-style, EOSS4) from the Chan Zuckerberg Initiative DAF, an advised fund of the Silicon Valley Community Foundation.