Sit back, relax and get ready to jet off on a whistle-stop tour of how we use optical character recognition. Since we’ve passed through security and seen how we safely handle applicants’ personal data during offer-to-arrival administration, it’s time to take a closer look at some of the technology that’s powering the student journey.
We’ll need to see a passport, but there won’t be any waiting around: our smart automation means your student jumps the queue and you get to your final destination much faster than with your existing processes.
As many of you will know, a passport is a key document that international applicants need to provide during the CAS and visa application process, and our software automates this request as part of the student journey. But CAS Shield goes much further, using optical character recognition and machine learning to verify that all of the personal details on the passport match up to the details in the student’s application: it is vital that this information is accurate during CAS and visa processing.
If you’ve passed through eGates during international travel, then you’ll be familiar with the concept of passports being checked and processed by machines in one way or another. In fact, machine-readable passports (MRPs) have been the norm in most countries for decades, and this means that the data on the identity page is encoded in optical character recognition format. This allows a computer to interpret the document and record the holder’s document type, name, nationality, date of birth, sex, and document expiration date.
But how does the process work?
First, CAS Shield prompts an applicant to upload an image of their passport page, which syncs to the student’s AWS S3 bucket in the cloud. Microservices written in Python are then used to crop, de-skew and scan the passport, ensuring that all the information is in the correct format. Finally, Python library ‘Python-Tesseract’ (a wrapper for Google's Tesseract-OCR Library), reads the characters in the passport’s Machine Readable Zone to produce the contents in machine-encoded text.
The results then get updated in the student’s record on CAS Shield, which determines if there are any conflicts with existing details. The system then flags up these conflicts with the user, ensuring that all personal information has been correctly filled out in the CAS process. Underpinning the whole system is a trained Machine Learning engine that is created by running it against many other passports.
So if you’re currently winging these processes with spreadsheets and email tools, CAS Shield offers you out-of-the-box functionality to improve the speed and accuracy of your CAS processing, leading to improved conversion rates and much better compliance outcomes.
Are you ready to streamline your CAS processes? Book a demo.
