foster/denoland-deno
1
0
Fork 0
mirror of https://github.com/denoland/deno.git synced 2025-01-21 04:52:26 -05:00
Code Issues Packages 1 Wiki Activity

update README

Browse source
This commit is contained in:
Hajime-san 2024-09-08 20:26:53 +09:00
parent 98e9304f2a
commit 4bc517eab2
1 changed files with 29 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

29
ext/canvas/README.md
View file

@ -1,3 +1,32 @@
# deno_canvas
Extension that implements various OffscreenCanvas related APIs.
## Image processing architecture in Rust
```mermaid
flowchart LR
Input["input binary<br/>( &[u8] )"]
II["intermediate image<br/>( DynamicImage )"]
Ops["processing pixel<br/>( ImageBuffer< P, S > )"]
Output["output binary<br/>( &[u8] )"]
Input --> II
II --> Ops --> II
II --> Output
```
The architecture of image processing in Rust is rely on the structure of
[image](https://github.com/image-rs/image) crate.\
If the input is a image of binary, it convert to an intermediate image
(`DynamicImage` in `image`) with using a decoder corresponding to its image
formats.\
After converting to an intermediate image, it can process various way for
example, to use the pixel processong operation
[imageops](https://github.com/image-rs/image?tab=readme-ov-file#image-processing-functions)
supplied by `image`.\
On the other hand, there can also to implement your own pixel processong
operation to refer to
[the implementation of imageops as here](https://github.com/image-rs/image/blob/4afe9572b5c867cf4d07cd88107e8c49354de9f3/src/imageops/colorops.rs#L156-L182)
or [image_ops.rs module](./image_ops.rs).\
You can treat any bit depth that supported by `image` with generics in the
processing pixel layer.