Image to text (OCR)
Recognise Simplified Chinese, Traditional Chinese, or English text locally and export readable text plus structured JSON with confidence data.
Remove backgrounds locally with a fast portrait model or a finer general model, then reuse the mask with transparency, a solid colour, or another image.
0 · 0 selected
No images selected yet.
This image’s crop has not been confirmed.
Select one or more sizes; each image is exported in every selected size.
One composition generates favicon, PWA and common app sizes, plus a manifest.
Base64 increases data size. This tool limits input to protect browser memory.
Remove backgrounds locally with a fast portrait model or a finer general model, then reuse the mask with transparency, a solid colour, or another image.
Choose the fast portrait or detailed general model, allowing the first download and integrity check to finish. Review the alpha mask, then reuse it for transparent, solid, or custom backgrounds; hair and translucent edges still need zoomed inspection.
The default result is a transparent PNG, or the same mask can be composited onto a solid or custom background. Zoom and compare before download.
Processing runs in your browser. Source images are not uploaded by this tool.
Your browser must decode images into memory. Very large files, long animations, and large batches may require substantial memory. Check the preview and export details, and verify the latest official requirements before submitting identity, application, or print-ready images.
Recognise Simplified Chinese, Traditional Chinese, or English text locally and export readable text plus structured JSON with confidence data.
Detect and refine page corners, correct perspective, apply colour, greyscale, or black-and-white enhancement, and optionally combine pages into a PDF.
Use local Real-ESRGAN inference to double image width and height; the verified model is cached after its first download and the page shows the current device limit.
Comments
0