Supported Documents

Atlas automatically classifies and extracts structured data from a wide range of documents used in financial workflows. You do not need to declare a document type at upload time — Atlas detects it from the image content and returns the appropriate fields in the document_type-specific payload within ocr_data. Each document type returns a distinct set of extracted fields, all following the same { "value": ..., "confidence_score": ... } structure. For the full list of fields per document type, see Supported Documents.

Document categories

Identity documents

These documents establish individual identity and are used for KYC verification in lending applications.

Document type	Description
`AADHAAR`	Indian national identity card. Extracts name, address, date of birth, and Aadhaar number. Includes a `redact_url` for a version of the document with the Aadhaar number masked — use this when storing or transmitting copies to comply with UIDAI guidelines.
`PAN`	Permanent Account Number card. Extracts name, father’s name, date of birth, and PAN number. Also returns `document_quality_flag` to identify whether the document is an `original`, `xerox_copy`, or `photo_of_photo`.
`DRIVING_LICENSE`	State-issued driving licence.
`VOTER_ID`	Election Commission voter identity card.
`PASSPORT`	Indian passport.

The document_quality_flag field (values: original, xerox_copy, photo_of_photo) is currently returned for PAN cards. Use it to flag low-quality copies for manual review before making credit decisions.

Business documents

These documents verify the legal registration and operational status of a business entity.

Document type	Description
`GST_CERT`	GST registration certificate. Extracts GST number, legal name, address, constitution of business, and validity dates.
`UDYAM_CERT`	Udyam registration certificate for MSMEs. Extracts registration number, enterprise name, owner name, address, type of enterprise, and NIC codes.
`SHOP_ACT_LICENSE`	Maharashtra Shops and Establishments Act licence. Extracts establishment name, employer name, address, nature of business, and total employees.
`FSSAI`	Food Safety and Standards Authority of India licence. Extracts registration number, business name, address, kind of business, and validity dates.
`DRUG_LICENSE`	Drug licence issued under the Drugs and Cosmetics Act.
`PARTNERSHIP_DEED`	Partnership agreement between partners. Extracts partner names and firm details.
`Memorandum of Association`	Memorandum of Association of a company.
`Articles of Association`	Articles of Association of a company.
`Certificate of Incorporation`	Certificate of Incorporation issued by MCA. Extracts company name, CIN, and incorporation date.
`BOARD_RESOLUTION`	Board resolution authorizing specific actions (e.g., loan acceptance).

Financial documents

These documents capture transaction and financing details for loan underwriting.

Document type	Description
`EBILL`	Electricity bill used for address verification. Extracts consumer name, address, and billing details.
`RTO_FORM_26`	Vehicle ownership certificate (Form 26).
`RTO_FORM_29`	Notice of transfer of ownership (Form 29).
`RTO_FORM_30`	Application for transfer of ownership (Form 30).
`RTO_FORM_34`	Hypothecation agreement form.
`RTO_FORM_35`	Termination of hypothecation form.
`RC_BOOK`	Vehicle registration certificate. Extracts owner name, registration number, engine and chassis details.
`INVOICE`	Dealer invoice for a product purchase. Extracts customer name, dealer details, product information, invoice amount, and hypothecation status.
`DELIVERY_ORDER`	Dealer delivery order. Extracts dealer name and address, customer name and delivery address, product details, serial/IMEI numbers, and EMI amount.
`BANK_STATEMENT`	Bank account statement. Extracts account number, IFSC code, bank name, account holder name, and statement period.
`MARGIN_MONEY_RECEIPT`	Receipt for margin/down payment. Extracts customer name, dealer name, product name, MMR amount, and hypothecation status.
`DEBIT_NOTE`	Dealer debit note for ancillary charges. Extracts customer name, insurance charges, RTO charges, accessory charges, and hypothecation charges.
`ACH_FORM`	Automated Clearing House mandate form.
`PURCHASE_ORDER`	Purchase order between buyer and seller. Extracts buyer/seller names, GSTINs, and total invoice amount.
`TAX_INVOICE`	GST tax invoice.
`INSURANCE`	Vehicle insurance policy. Extracts insured name, insurer, policy number, engine and chassis numbers, IDV, gross premium, tenure, and hypothecation status.
`ITR`	Income Tax Return document. Extracts PAN, total income, and filing details.
`SALARY_SLIP`	Salary slip issued by employer. Extracts employee name, employer name, net salary, and pay period.

Property documents

These documents are used in property loan and mortgage underwriting. Many are Maharashtra-specific revenue and registration records.

Document type	Description
`SALE_DEED`	Registered sale deed for property transfer.
`INDEX_2`	Index II extract from the Sub-Registrar’s office. Extracts owner names, mortgagor name, loan amount, property valuation, survey numbers, property area, and mortgage dates.
`SATBARA`	7/12 extract (Satbara Utara) from Maharashtra land records.
`RENT_AGREEMENT`	Rental or leave-and-licence agreement.
`GIFT_DEED`	Registered gift deed.
`NAMUNA_8`	Namuna 8 form from Gujarat revenue records.
`CHATUSIMA`	Chatusima boundary document for land parcels.
`GAVTHAN_CERTIFICATE`	Gavthan certificate for village settlement land.
`PROPERTY_TAX_RECEIPT`	Municipal property tax payment receipt.
`PROPERTY_CARD`	Property card from urban land records.
`GUNTHEWARI_CERTIFICATE`	Gunthewari regularisation certificate.
`MEMORANDUM_OF_DEPOSIT_OF_TITLE_DEEDS`	Memorandum of deposit of title deeds (MODT) for equitable mortgage.

Asset and image documents

These document types process images rather than scanned paperwork, and are used to verify physical asset condition and identity at the point of delivery.

Document type	Description
`IMEI_CLOSEUP_IMAGE`	Close-up photo of a device’s IMEI/serial sticker. Extracts model, RAM, ROM, IMEI numbers, product SKU, price, and serial number.
`ASSET_OPEN_BOX_IMAGE`	Photo of an open-box asset. Detects whether the customer is present (`customer_present`) and counts the number of visible faces (`number_of_faces_visible`).
`AADHAR_EXTRACT_IMAGE`	Cropped Aadhaar photo extract. Detects whether a photo is present (`photo_present`). Used to verify that the physical Aadhaar card contains a photograph.
`PRODUCT_STICKER`	Product label or sticker image.
`GOLD_PLEDGE_CARDS`	Gold pledge card issued by lender. Extracts customer name, pledged asset details, and loan amount.
`GAS_BILL`	Gas utility bill used for address verification.
`PHONE_BILL`	Telephone bill used for address verification.
`WIFI_BILL`	Internet service bill used for address verification.

Type-specific OCR fields

Each document_type in the ocr_data array returns a different field schema. For example, an AADHAAR document returns name, address, aadhaar_number, and date_of_birth, while an INVOICE returns customer_name, dealer_address, total_invoice_amount, and so on. Your processing logic should branch on document_type to read the correct fields.

{
  "ocr_data": [
    {
      "document_id": "doc-001",
      "error_code": "",
      "error_reason": "",
      "data": {
        "document_type": "PAN",
        "name_as_per_pan": { "value": "D MANIKANDAN DURAISAMY", "confidence_score": 0.95 },
        "pan_number": { "value": "PLZPM5601F", "confidence_score": 0.99 },
        "dob": { "value": "16/07/1986", "confidence_score": 0.98 },
        "document_quality_flag": { "value": "original", "confidence_score": 0.97 }
      }
    }
  ]
}

Extraction failures

If Atlas cannot extract data from a document, the data object will be empty or partially populated, and the error_code and error_reason fields will contain a machine-readable code and a human-readable explanation.

{
  "document_id": "doc-002",
  "error_code": "UNRECOGNISED",
  "error_reason": "document not in list",
  "data": {}
}

Common error codes include UNRECOGNISED (document type not supported) and INVALID_DOC (document could not be processed, for example due to image quality or multiple subjects in frame). For details on every extracted field per document type, see Supported Documents.

Get Started

Core Concepts

Guides

Resources

Document categories

Type-specific OCR fields

Extraction failures

​Document categories

​Type-specific OCR fields

​Extraction failures

Document categories

Type-specific OCR fields

Extraction failures