This is a library focused on bringing dithering to Swift since the CIFilter for dither is mostly noise. The implementation is done via the Accelerate framework.
- Bayer 8x8:
- Photo by Andraz Lazic on Unsplash
-
Floyd-Steinberg:
- 1-bit:
- 2-bits:
To use the library you will have some access points based on UIImages and CIFilter
CPU:
import SwiftDithering
function yourFunction() {
//Highly Recommended
Task{
do{
let ditheredImage = try image?.applyErrorDifusion(withType: .floydSteinberg)
//or
//try image?.applyOrderedDither(withSize: .bayer8x8)
//or
//try image?.applyThreshold(withType: .fixed)
}catch{
//handle errors here
}
}
}the functions also have more accessible parameters, but for simplicity, this example does not, For more details check the Sample/DitherTester/DitherControl and go to any of the retrivedDitheredImages
GPU:
import SwiftDithering
function yourFunction() {
let filter = OrderedDithering()
filter.setValuesForKeys([
"inputImage": inputCIImage,
"ditherType": ditherType.getCIFilterID(),
...
])
return filter?.outputImage
}I recommend transforming the outputImage to CG before using in IOS, same thing for the respective variants on mac.
Since is my first time doing something like this a lot of the code is prone to change, I will try to do it in the most non-disruptive way but I cannot guarantee it will work in every single version.
Currently Features:
- Bayer matrix;
- Floyd Steinberg Error Diffusion;
- Stucky Error Difusion.
- Fixed Threshold
- Random Threshold
- Uniform Threshold
- Clustered Dots
- Central White point
- Balanced Centered point
- Diagonal clustered
- All Ordered Dithers are Now CPU and GPU supported.
- The CPU variants are calculated as sequential and are expensive so tasks and loadings are needed for the user UI.
- Feel free to suggest performance improvements and report any bug that you may find while using this lib
- These dithers are based on the matlab article stored in the archive.today and other research