Image to Text — OCR with Bold, Italic & Color Detection
Browser-OnlyExtract text from images and PDFs while preserving bold, italic, font size, and color. Visual overlay on the original document. 100% browser-based — your file never leaves your device.
Drop PDF or image here, or click to browse
PDF, PNG, JPEG, WebP — processed entirely in your browser
How to Use
Step-by-step guide to get the most from this tool
- 1
Upload file
Drop or select a PDF or image (PNG, JPEG, WebP).
- 2
Wait for extraction
PDF text layers extract instantly; images and scanned PDFs run OCR with a progress bar.
- 3
Review in Overlay
See bounding boxes over the original — bold, italic, and color-coded.
- 4
Edit and export
Correct OCR errors in Structured view, then copy or download as text, HTML, Markdown, or JSON.
Features
What makes this tool stand out
Dual engines
PDF.js for text PDFs; Tesseract OCR for images and scans.
Formatting detection
Bold, italic, font size, and color per word.
Visual overlay
Bounding boxes aligned to source document.
Multiple exports
Plain text, HTML, Markdown, and JSON.
Browser-only
Nothing uploaded — PDF.js and Tesseract run locally.
Free
No signup required.
Related Tools
More free utilities you might find useful
Frequently Asked Questions
Quick answers to common questions
How accurate is the bold/italic detection?+
For text-based PDFs, bold and italic are read directly from the PDF's font data — this is exact, not a guess. For images and scanned documents, Tesseract's OCR engine estimates bold/italic based on stroke width and character shape analysis — accuracy is typically 85–95% on clear, high-resolution images and lower on blurry or low-quality scans.
How is text color detected?+
For PDFs with a text layer, ToolReign uses a neutral default in v1 (PDF.js does not expose fill color via getTextContent). For images, color is sampled directly from the darkest pixels within each detected word's bounding box — this works reliably for standard dark-text-on-light-background documents but may be less accurate for stylized or low-contrast text.
Does this tool upload my images or documents anywhere?+
No. Both the PDF reading engine (PDF.js) and the OCR engine (Tesseract.js) run entirely as WebAssembly/JavaScript inside your browser. Your file never leaves your device.
Can I extract text from scanned PDFs?+
Yes. When a PDF has no usable text layer, ToolReign automatically renders each page and runs OCR, with a notice that scanned mode is active.
What export formats are supported?+
Plain text, HTML with inline styles, Markdown with heading heuristics, and full JSON of the extraction data structure.
Is there a file size limit?+
Images over 10 MB show a warning; images over 25 MB are blocked. PDFs with more than 20 pages show a warning — pages are processed on demand.
Your data is processed entirely in your browser. Nothing is sent to any server.