Skip to content

A lightweight desktop utility for Windows that periodically captures a specific region of your screen, runs OCR and triggers automated actions when specific text is detected.

License

Notifications You must be signed in to change notification settings

sobattry/OCR-Alarm

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Screen OCR Monitor

A lightweight, modern desktop utility for Windows that periodically captures a specific region of your screen, runs deep-learning-based OCR (Optical Character Recognition), and triggers automated actions when specific text is detected.

Built with Python, PyQt6, and RapidOCR.

Features

  • Real-time Monitoring: Captures a screen region at a user-defined frequency (e.g., every 2 seconds).
  • Deep Learning OCR: Uses RapidOCR (ONNX) for high-accuracy text recognition, far superior to standard Tesseract for screen text.
  • Smart Filtering:
    • Keyword Detection: Trigger actions only when specific text is found.
    • Fuzzy Matching: Detects keywords even if there are small OCR errors (e.g., "Login" vs "Log1n").
    • Change Detection: Pauses OCR if the screen hasn't changed to save CPU.
  • Automated Actions:
    • Telegram: Send the detected text and/or a screenshot to your Telegram chat.
    • Email: Send an email alert with the text and screenshot attachment.
    • Local Logging: Save detection history to a JSON log file.
  • Modern GUI:
    • Dark-themed, minimal interface.
    • Visual "Select Region" overlay tool.
    • Configurable options (stop on match, include/exclude screenshots).

Installation

Prerequisites

  • Python 3.8 or higher.

About

A lightweight desktop utility for Windows that periodically captures a specific region of your screen, runs OCR and triggers automated actions when specific text is detected.

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages