Digitization & Cognitive Process Automation for Cheque Truncation System (CTS)
Using proprietary computer vision, computational geometry and machine learning algorithms, SnapChek automates the extraction and validation of information on cheques at 100% accuracy.
Using Computer Machine Reading, Snapchek can accurately extract the following data from a cheque - date, name, amount in words / figures, account number, MICR, IFSC code. Data can be printed or hand-written, as long as it is in English.
The most striking feature of SnapChek is the Automatic Signature Verification which matches signatures on the cheque to the ones the core banking database at high levels of accuracy.
How SnapChek works
Scanning
It can process any cheque scanned with a standard cheque scanner, commonly used by banks.
Scanning
Identification
Its region-of-interest detector identifies each element — date, name, amount, signature etc.
Identification
Segregation
Based on scan quality, it segregates the cheques into processable and non-processable streams. The latter is directly sent for human intervention.
Segregation
Extraction
In the processable stream, SnapChek extracts date, amount in words / figures, signature, account number, MICR and IFSC code.
Extraction
Validation
It then validates the data — date is within 90 days, amount in words and figures are the same, signature on cheque matches signature on file etc.
Validation
Output
Based on this, it approves, rejects or flags for human intervention.
Output
SnapChek Brochure download
For a more detailed look into the product, features and processes, download our brochure here.
How we achieve near-100% accuracy with SnapChek
Multiple algorithms for region-of-interest detection
In SnapChek, there are multiple algorithms working simultaneously on region-of-interest identification, ensuring 100% detection.
Ensemble of algorithms for signature verification
SnapCheks algorithms are tuned to account for challenges of low image quality, change over time, ambient noise etc. while verifying signatures with the database.
Adaptive field-of-view
Before extraction, SnapChek accommodates inconsistencies in data placement, such as printing the date slightly outside its boxes.
Dual algorithmic journeys for building quorum
It builds quorum using two algorithms for every field. Each algorithm has different deep learning bases and maths for feature extraction methods, number of layers, loss functions etc.
De-noising
De-noising systems from simple Otsu methods to deep learning-based segmentation algorithms improve SnapChek’s extraction quality.
Claim your POC
Take SnapChek for a spin with our customised proof-of-concept for banks and financial institutions in India. Fill the form below to claim yours now.
Claim your POC
Take SnapChek for a spin with our customised proof-of-concept for banks and financial institutions in India. Fill the form below to claim yours now.
