TurboLens

TurboLens positions itself as a screening layer that sits in front of high-trust decisions. Instead of treating OCR as the only step, it asks a prior question: is the submitted image trustworthy? The forgery detection service analyzes pixel structure for splicing, copy-move edits, retouching, and altered fields, returning a real-or-forged verdict plus the proportion of tampered pixels, suspicious-region coordinates, and a heatmap image. A second, early-access service screens whether an image was created or modified with generative AI, producing a three-state result and an AI probability score. The two signals stay deliberately separate so that teams route each kind of risk through the workflow it deserves.

Once authenticity is screened, TurboLens switches to document intelligence. Its OCR and structured extraction capabilities are aimed at complex Southeast Asian documents: regional invoices and receipts, national IDs and passports, multilingual scans, contracts, tax forms, and customs paperwork. Output is structured JSON with configured field groupings, so results can be mapped directly into a product schema and pushed into downstream systems such as KYC checks, underwriting, claims triage, or case-management tools.

Integration is intentionally light. A single REST request submits a document page or image; the API returns fraud and screening signals with evidence references; and clean files continue automatically while flagged ones go to review. TurboLens advertises a free sandbox and free-trial entry point for testing your own samples, with enterprise arrangements and custom integrations handled through direct sales. Because the platform is API-first, it complements rather than replaces existing OCR, face matching, watchlist screening, and reviewer tooling.