
Handwriting Recognition-Based Correct Answer Identifier
Software as a Service
NoCategory
AI & Machine LearningClients
Techstack
Purpose
Grading paper exams by hand is slow and error-prone, especially when tests mix handwritten identification answers with shaded multiple-choice bubbles. This system automates both, scoring a captured answer sheet against its answer key in a single pass.
Description
A PyQt5 desktop application captures each answer sheet through a camera and reads a QR code to load the matching sheet format. OpenCV isolates the paper and each answer region, Microsoft's TrOCR handwriting-recognition transformer transcribes the written answers, and an optical mark recognition routine detects shaded bubbles. Recognized answers are compared against the key and results are exported to spreadsheets, with the tool packaged as a standalone Windows executable via PyInstaller.




