NAPS2 failed to preserve text in imported PDF after applying image operations

[zephyros-dev]

Describe the bug I'm planning on using NAPS2 for cleaning up pages of scanned Internet Archive books to make it more readable. The pages in those books have already been OCR'ed and can be used directly in the file. However, when I used the image operations (I tried Black and White, Sharpness, Document correction options) on a page and save it to PDF, the existing text is lost. This does not happen if I just saved the page directly without doing the image operation. Also oddly enough when I tested NAPS2 on Mac ARM version installed from brew, it managed to preserved the text even after the image operations, which is what I wanted to do. Since it works normally on Mac I think this can be qualified as a bug.

To Reproduce Steps to reproduce the behavior:

1. Import the attached test.pdf file into NAPS2
2. Do image operations on some pages (Black and White, Sharpen, Document Correction)
3. Click Save PDF

Expected behavior The saved files with image operations preserve the existing text in the imported PDF

Screenshots

Desktop (please complete the following information):

• OS: Windows 10, Fedora 41
• Version: 7.5.2, 8.0b1 (Neither version works on these OSes)

Additional context This works normally on Mac M1 Sequoia 15 with NAPS 7.5.2