Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Updates to hls::vision library: #16

Open
wants to merge 12 commits into
base: main
Choose a base branch
from
Open

Updates to hls::vision library: #16

wants to merge 12 commits into from

Conversation

ManuelSaldana
Copy link
Collaborator

  • Default value (GBRG) for bayer format argument was removed from DeBayer function. Now users need to explicitly pass the required format. This will hopefully prevent unexpected mismatches.

  • Added bayer format argument to RGBBayer function.

  • Added new function BGR2Bayer to make it easier to just process images loaded with OpenCV without having to convert the color.

  • Updated Bayer-DeBayer example using the above changes.

  • Updated golden images for Bayer and Debayer test.

  • Added two new variations to TransformPixel function. One where the "ij" pixel coordinates is passed to the functor argument, and the other where an "enable" argument is passed to the functor argument. This will provide more info to process pixels.

  • Added new method ConvertTo() the Img class. This is similar to OpenCV's function to scale and add factors to pixels of an image.

  • Add .gitignore

- Default value (GBRG) for bayer format argument was removed from DeBayer function. Now users need to explicitly pass the required format. This will hopefully prevent unexpected mismatches.
- Added bayer format argument to RGBBayer function.
- Added new function BGR2Bayer to make it easier to just process images loaded with OpenCV without having to convert the color.
- Updated Bayer-DeBayer example using the above changes.
- Updated golden images for Bayer and Debayer test.

- Added two new variations to TransformPixel function. One where the  "ij" pixel coordinates is passed to the functor argument, and the other where an "enable" argument is passed to the functor argument.   This will provide more info to process pixels.

- Added new method ConvertTo() the Img class. This is similar to OpenCV's function to scale and add factors to pixels of an image.

- Add .gitignore
- Also removed unnecessary ij variables in TransformPixel function.
ManuelSaldana and others added 10 commits May 27, 2024 16:27
- Also added the example designs for testing each function.
- For the GammaCorrection function, a table generator has been added. This is a temporary solution until we can generate tables at compile time.
- Also updated the .gitignore file
The demo now has a PASS/FAIL check against a golden image. Even though the demo is meant to be used for video, a test with a single image can at least check if the verilog code would work for one frame.

The check also has a flag (SMALL) to select a small image for quick cosim testing. If this flag is not set (Default) a 4K image is used.
Makefile has been updated and run_libero.ps1 has been added
- Updated readme
- Updated .hex file with additional printout
- Modified gdb.txt to load the .elf, start the execution and quit.
- SW makefile now always updates the .hex file that Libero uses to
  compile the bitstream
The update includes the commands to:

1. Compile the SmartHLS module and generate Verilog and the C++ API
2. Compile the software for the Soft-MiV RV32 CPU
3. Compile the Libero project
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants