Errors
The table below lists the main error codes and checks that may occur during KYC procedures on the NVGlobal platform. For convenience, a concise description, user comment, and a brief explanation of the technology triggered behind each check are included. This reference guide is aligned with the internal «Errors» section and response examples of the /kyc/process and /kyc/session/status methods.
| No. | Verification code | Description | Comment | Technology description |
|---|---|---|---|---|
| 1 | [general] bad image | The uploaded file is not an image | You sent a file that is not an image | When uploading an image, the system checks its type. Only JPEG and PNG are supported. If the file does not match these formats, an error is returned indicating that the uploaded file is not an image. |
| 2 | [general] too small image | The image is too small for successful recognition | The image resolution is too low for successful processing; please take a higher-quality photo | The system checks the resolution of the uploaded image in pixels. For successful processing, the larger dimension (width or height) must be at least 300 pixels. If this condition is not met, an error is returned recommending to upload a higher-quality photo. |
| 3 | [general] no documents on image | No documents found in the image | Make sure you are taking a photo of the required document | The system uses a neural network (detector) to analyze the image. If no objects that can be identified as documents are detected in the image, an error is returned recommending to make sure the required document is present in the photo. |
| 4 | [general] not a selfie | No faces detected outside the document | Wipe the camera and try taking a new photo, make sure the image is in focus | The system analyzes the image, checking for the presence of a face outside the document. A face detector is used to identify a face of a certain size. If no face is detected outside the document, an error is returned recommending to make sure the image is in focus and the face is clearly visible. |
| 5 | [general] low light | Insufficient lighting (image is too dark) | Use additional light sources or choose a better-lit place for shooting | The system applies an algorithm to analyze the lighting level of the image. If the lighting level is below the set threshold, an error is returned recommending to use additional light sources or choose a better-lit place for shooting. |
| 6 | [general] age requirement mismatch | Age does not meet the requirements | Your age does not meet the requirements, please return later | The system uses a neural network to determine a person’s age from their face image. The age is determined with an accuracy of ±4 years. If the calculated age is below the set threshold or differs by ±4 years from the age of the person in the document, an error is returned indicating that the age does not meet the requirements. |
| 7 | [general] black and white image | Black and white image | Do not use filters when taking photos; the image must be in color | The system analyzes the color channels of the image. If the image contains only one channel (black and white), an error is returned recommending to use a color image without filters. |
| 8 | [general] too many faces | More than one face found in the document image | Make sure you are photographing only the document and there are no other faces in the background | The system uses a face detector to analyze the image. For a document, it verifies that no more than the appropriate number of faces for that document type are present in the photo. If additional faces are detected in the document image, an error is returned. |
| 9 | [document] unknown document type | Unknown document type in the image | Make sure you are taking a photo of the required document | The system uses a document detector to analyze the image. If a document is detected in the photo but its type cannot be identified (it is missing from the database of known document types), an error is returned recommending to make sure the required document is being photographed. |
| 10 | [document] bad document type | Incorrect document type | Make sure you are taking a photo of the required document | The system uses a document detector to analyze the image. If the document is successfully classified but does not meet the requirements of the current scenario at this step, an error is returned recommending to make sure the required document is being photographed. |
| 11 | [document] too small image | The document is too small | The document in the image is too small, bring the camera closer | The system checks the size of the document in the image recognized by the detector. If the larger side of the document (width or height) is less than 200 pixels, this is considered insufficient for successful text recognition. An error is returned recommending to bring the camera closer. |
| 12 | [document] bad pages | Problem with pages | Make sure you are photographing the required pages of the document | The system classifies the recognized document and checks if the page spread matches the scenario requirements. If another page of the document is uploaded instead of the required spread (for example, the main spread of a passport), an error is returned recommending to photograph the correct pages of the document. |
| 13 | [document] no face | No faces found in the document image | Make sure the photo in the document is visible | The system uses a face detector to analyze the document image. If no face (the passport holder’s photo) is detected in the document image, or if it is not visible enough for successful recognition, an error is returned recommending to make sure the photo in the document is clearly visible. |
| 14 | [document] text fields not visible | Fields are not visible | Wipe the camera and try taking a new photo, make sure the image is in focus | The system uses a field detector to analyze the document image. If a required field is not detected (for example, due to blur, hand coverage, or other factors), an error is returned recommending to take a new photo, making sure the image is in focus and the document fields are visible. |
| 15 | [document] bad mrz checksum | MRZ error | Wipe the camera and try taking a new photo, make sure the image is in focus | The system uses OCR (optical character recognition) technology to extract data from all document fields, including the machine-readable zone (MRZ) lines. Based on the extracted data, checksums are calculated according to the legally established algorithm. If the checksums do not match those specified in the MRZ, or if the MRZ lines are not fully visible, an error is returned recommending to take a new photo, making sure the image is in focus. |
| 16 | [fraud] display or screen | Photo of a screen/display | A screen copy cannot be used; please take a photo of the original document | The system applies a neural network classifier to analyze the image. The neural network calculates the probability that the photo contains signs of a screen image. If the probability is high, an error is returned recommending to take a photo of the original document. |
| 17 | [fraud] image edited | The document image shows signs of graphic editor tampering | An edited photo cannot be used; please take a photo of the original document | The system analyzes the metadata of the uploaded image. If the metadata contains information indicating that the file was opened or modified in a graphic editor, an error is returned recommending to take a photo of the original document without editing. |
| 18 | [fraud] image printed | The document image shows signs of a photocopy | A photo of a printout cannot be used; please take a photo of the original document | The system applies a neural network classifier to analyze the image. The classifier determines the probability that the image is a photo of a printed document, based on signs of black-and-white or color printing. If the probability exceeds the set threshold, an error is returned recommending to take a photo of the original document. |
| 19 | [face] bad face matching | Faces in the document and the selfie image do not match | Take a selfie holding the document; both your face and the document must be fully visible | A pairwise comparison of faces in the documents and the selfie with the document is performed. If the number of faces is N, the number of matching pairs must be N−1. If at least one pair does not match, an error is returned. |
| 20 | [face] not all faces found | Not all faces required for verification were found | Wipe the camera and try taking a new photo, make sure the image is in focus | During the Liveness procedure, 5 photos are taken for verification. If a face is not detected in at least one photo, an error is returned recommending to make sure the camera is in focus. |
| 21 | [fraud] black list | User is blacklisted | You did not pass the verification | A search for the user’s face is performed in the blacklist. In case of a match, an error is returned indicating that the user is blacklisted. |
| 22 | [legal] legal proceeding existence | The user has an active legal proceeding | You did not pass the verification | The system queries the database of the Federal Bailiff Service (FSSP RF). If an active legal proceeding is found for the user, an error is returned indicating that the verification failed. |
| 23 | [legal] document not valid | The document is invalid | Your document is invalid; please replace it and try again | The system queries the Ministry of Internal Affairs (MVD) database to check the document status. If the document is found to be invalid, an error is returned recommending to replace it with a valid one and retry the verification. |
| 24 | [system] balance too low | Insufficient balance | Top up your balance and try again | The system checks the user’s current account balance. If funds are insufficient to complete the request or operation, an error is returned recommending to top up the balance and try again. |
| 25 | [system] exception | Internal service error | Please try sending the image again | The error is returned if the system encounters an unexpected internal error during data processing. The user is prompted to try sending the image again. |
| 26 | [document] bad quality | Poor image quality | Wipe the camera and try taking a new photo, make sure the image is in focus | The system applies an image quality assessment algorithm. If the image is of low quality, insufficient for successful recognition, an error is returned recommending to wipe the camera and take a new photo. |
| 27 | [general] too many tries | Maximum number of attempts exceeded | You have exhausted the allowed number of attempts to upload the expected image type | The system limits the number of upload attempts for each document type to 6 within a single session. If the user exhausts the allowed number of attempts for any of the supported document types, the session is blocked, and further uploading becomes impossible. |
| 28 | [fraud] manipulation detected | Data manipulation attempt detected | Please retake the photo in accordance with the requirements | The system tracks the sequence of uploaded documents within a single session. If the use of different documents (e.g., documents of different people) or documents of the same user but of different types (e.g., passport and driver’s license) is detected, an error is returned recommending to retake the photo in accordance with the requirements. |
| 29 | [fraud] document modified | The document is physically modified | The document must not have physical modifications or pasted elements | The system analyzes the document image for signs of physical modifications. It checks for pen corrections, as well as pasted elements, such as printed and glued fields. If such modifications are detected, an error is returned recommending to use the original and unmodified document. |
| 30 | [fraud] logical mismatch | Logical inconsistencies detected | You did not pass the verification | The system performs a check of the document data for logical inconsistencies. Dates, series, and document numbers are analyzed. If obvious errors are detected, such as non-existent dates or serial numbers, an error is returned indicating that the verification failed. |
| 31 | [fraud] bad document photo | The photo in the document looks suspicious | Do not cover the photo and the area around it with fingers or foreign objects | The system analyzes the photo in the document for signs of forgery. If it is suspected that the photo is pasted over the original or drawn in a graphic editor, an error is returned. |
| 32 | [fraud] mobile device | The photo was taken from a mobile device screen | A screen copy cannot be used; please take a photo of the original document | The system uses a neural network (detector) to analyze the image. The detector looks for signs of devices with screens (smartphones, monitors, TVs, etc.) on which a document or selfie might be displayed. If such a device is detected, an error is returned recommending not to photograph such devices. |
| 33 | [fraud] document damaged | The document is damaged | The document is invalid; it must be replaced | The system applies a neural network to analyze the document image for damage. If the document has obvious damage (torn or ripped pages, ink stains, or other defects), an error is returned recommending to replace the damaged document. |
| 34 | [document] cropped | The document is not fully visible | The document must be fully visible; hold the camera further from the document | The system checks whether the document is fully visible in the photo after its detection. If part of the document goes beyond the image boundaries and is not captured in the frame, an error is returned recommending to take a new shot so that the document is fully visible. |
| 35 | [document] no signature | The document is missing a signature | The document is invalid; it must be replaced | The system uses a neural network (detector) to search for signatures in the document. If the signatures of the document owner or the issuing authority are not detected, an error is returned recommending to replace the document with a valid one. |
| 36 | [document] no stamp | The document is missing a stamp | The document is invalid; it must be replaced | The system applies a neural network (detector) to analyze the document image. If the required stamp is not detected, an error is returned recommending to replace the document with a valid one. |
| 37 | [document] no required objects | The document is missing required elements | The document is invalid; it must be replaced | The system analyzes the document for the presence of all mandatory elements (stickers, holograms, and other security features). If any of these elements are missing, an error is returned recommending to replace the document with a valid one. |
| 38 | [document] text field has low confidence | The field was recognized with low confidence | Wipe the camera and try taking a new photo, make sure the image is in focus | The system uses a field detector to analyze the document image. If a required field is not detected (for example, due to blur, hand coverage, or other factors), an error is returned recommending to take a new photo, making sure the image is in focus and the document fields are visible. |
| 39 | [fraud] multiple documents | Different documents of the same type | You did not pass the verification | The system analyzes the uploaded documents and tracks changes in them within a single session. If the user submits the same document type but with different data (for example, first uploads one passport and then another, or modifies information in the document), the system records a manipulation attempt. A corresponding error is returned with a rejection of the verification. |
| 40 | [document] expired | The document has expired | The document is invalid; it must be replaced | The system checks the expiration date of the uploaded document. If the document is expired and is not valid according to the established requirements (for example, the passport or driver’s license has expired), a corresponding error is returned. The user is recommended to upload a valid document. |
| 41 | [system] service timeout | Service response timeout exceeded | Please try sending the image again | The error is returned if the system encounters a microservice response timeout during data processing. The user is prompted to try sending the image again. |
| 42 | [document] other error | The document does not meet the requirements | Please present another document | The error is specifically designed for modes where we want to block a document without providing a detailed hint about what exactly is unsatisfactory. It is used to prevent sophisticated fraud, mostly as a temporary solution. |
| 43 | [face] is deepfake | The face in the image shows signs of generation or spoofing | Use an original photo with a real face | The system analyzes the face and detects signs of generation. If the probability exceeds the allowed threshold, an error is returned recommending to upload an authentic photo. |
| 44 | [face] is not alive | The image failed the liveness check | Provide a photo of a living person taken in good lighting | The system analyzes the face in the image and detects signs of a lack of liveness: skin staticity, texture, sharpness, reflections, and other parameters. If the cumulative score indicates a high probability of a «non-live» face (mannequin, photo, screen), an error is returned requiring the image to be replaced. |
| 45 | [document] no required fields | The document is missing required fields | The document is invalid; it must be replaced | The system analyzes the document for the presence of all mandatory text fields. If any of these fields are missing, an error is returned recommending to replace the document with a valid one. |
