How it works
Complete the job in three clear steps
- Choose an owned JPG, PNG, WebP, or BMP image.
- Set the available options and create an inpainted PNG and detected-word report.
- Review and download or copy the result before leaving the page.
Private Lucaden images & media tool
Remove Text from Image
Detect printed text locally with OCR, reconstruct the selected word regions with OpenCV inpainting, and download a separate PNG.
Free to use Your image stays in this browser. Public OCR language files may download on first use, then word detection, masking, inpainting, and PNG export run locally.
Loading tool workspace…
Important
Check the result before relying on it
Keep an untouched copy of important source files. Browser processing depends on available device memory, and you should inspect every downloaded result before deleting the original.
Practical details
What this Lucaden tool does
Tesseract.js proposes word boxes and confidence scores, then OpenCV masks and inpaints those regions from nearby pixels. This deterministic workflow works best on small text over simple or repeating backgrounds; it is not generative scene reconstruction.
The original stays unchanged; Lucaden creates a separate result in browser memory.
Included capabilities
- Local OCR word detection
- Confidence and padding controls
- Telea or Navier-Stokes inpainting
- CSV region report
Useful for
- Clean labels from your own mockup
- Remove a date from an authorized scan
- Prepare a reusable background
- Repair text-marked design drafts
Questions
About this tool
Will every word disappear cleanly?
No. OCR can miss stylized, curved, low-contrast, rotated, or handwritten text, while inpainting can smear complex backgrounds. Review every region and keep the source.
Can I remove a watermark or ownership mark?
Do not remove copyright, ownership, provenance, safety, or authenticity marks unless you are the rights holder and the change is lawful and non-deceptive.
Open-source foundations
Built transparently
Lucaden documents its standards sources, architectural references, and permissive runtime dependencies while keeping this workspace consistent with the Lucaden interface.