When Images Speak: In Pixels Text Extraction and Verification Using OCR & AI

GDG on Campus Sudan University of Science & Technology - Khartoum, Sudan

This project demonstrates a basic identity verification system using OCR to extract handwritten text from images and match it with user input for access control.

May 22, 7:00 – 9:30 PM (UTC)

12 RSVP'd

RSVP

Key Themes

AIBuild with AIMachine Learning

About this event


This project showcases a simple identity verification system based on Optical Character Recognition (OCR), aimed at extracting handwritten text—like names or signatures—from an image and comparing it with user-provided input. Built in a Google Colab environment, the system utilizes Tesseract OCR and image processing libraries such as OpenCV. Images are loaded and preprocessed through grayscale conversion and thresholding to enhance OCR accuracy. The extracted text is cleaned and normalized, then compared with the normalized user input. Access is granted only if a match is found. This project is ideal for basic applications in digital signature verification, identity checks, or secure login systems using scanned documents.

Speakers

  • Nuha Jadu

    Creator at Google IO Amsterdam

  • Beyza Nur Elitok

    Google Developer Groups on Campus Organizer

Organizers

  • Hiba Eljozouly

    Organizer

  • Dr. Anwar Dafa-Alla

    Sudanese Researchers Foundation (SRF)

    Chapter Supervisor

Contact Us