AI Document Scanner

AI Document Scanner turns paper documents and selected photos into organized PDF files with optional AI-powered text recognition. It is designed for receipts, contracts, invoices, forms, notes, letters, study material, and other pages that you need to capture, clean up, search, and share. The app combines Android camera capture, local image processing, a private on-device document library, and a user-configured Vision API connection in one focused workflow.

Start by scanning a page with the in-app camera or selecting existing images through Android Photo Picker. Smart capture can help you frame a stable page, while manual capture remains available whenever you want direct control. After capture, review the detected page edges, adjust all four corners, rotate the page, and choose the visual treatment that works best for the source. Color, grayscale, black-and-white, and original-image options make it easy to prepare everything from colorful receipts to high-contrast text pages.

Optional AI document analysis can recognize visible text, classify the document, check page quality, suggest a useful file name, and help identify the physical page boundary. High Accuracy mode performs an additional overlapping-region OCR pass to inspect small print, edge text, table cells, stamps, labels, and low-contrast details. AI recognition is handled by the Vision-capable API endpoint configured on the device. The app does not operate a developer-owned document-processing server. You control whether AI is enabled and which compatible provider receives the temporary page copy.

Choose the PDF result that matches your task. Text Only creates clean white pages containing selectable AI-recognized text, without retaining the photographed colors or background. Image + Text preserves the processed page image and adds an invisible searchable text layer so the document keeps its original appearance while supporting search and text extraction. Compact, Balanced, and Ultra quality settings provide practical control over output size and image detail. PDF generation, image placement, text-layer creation, file naming, and document storage happen locally on the Android device.

Multi-page projects are easy to manage before export. Add more pages, replace a page, delete unwanted pages, rotate or reprocess individual pages, and press and hold to reorder the final sequence. A clear generation screen reports AI waiting, image loading, compression, searchable-text layering, page completion, PDF saving, and final library registration. This makes longer Image + Text exports easier to follow instead of leaving an unmoving indicator on screen.

Saved documents appear in the local library with a thumbnail, page count, file size, and recognition status. Open a PDF in the built-in viewer, search recognized text across saved documents, rename files, inspect document details, share a PDF through Android's standard share sheet, or delete it when it is no longer needed. Text Only PDFs also provide a selectable text view for long-press selection and copying. No account or sign-in is required.

Five interface styles are included and can be changed at any time in Settings: Vanta Scan, Paperlight, Mono Grid, Luma Flow, and SnapFlow. The selected style is saved on the device and remains active after the app is restarted. Camera guides, smart auto-capture, haptic feedback, reduced motion, default PDF quality, OCR accuracy, and the default PDF content mode can also be adjusted.

AI can be disabled at any time. Camera scanning, manual crop, perspective correction, local filters, page organization, normal image PDF creation, viewing, sharing, and deletion remain available without AI. Recognition quality depends on page focus, lighting, language, layout, handwriting, and the capabilities of the configured Vision model, so important results should always be reviewed before use.