Text extraction from a 16-segment display using computer vision techniques.
Text_Detector
is designed to detect and extract text from videos and images, especially from a 16-segment display. The project utilizes OpenCV for image and video processing and EasyOCR for text recognition. A Flask-based UI is also provided for visualization.
- Make sure Python 3.x is installed.
- Clone the repository:
git clone https://github.com/Dv04/Text_Detector
cd Text_Detector
- Install the required libraries:
pip install -r requirements.txt
- Extract frames from video:
python ComputerVision/CameraVision.py
- Detect text from video:
python ComputerVision/ComputerVision.py
- Detect and test text extraction from image:
python ComputerVision/ImageVision.py
- Launch the Flask UI:
python ui/app.py
Then access the UI at http://127.0.0.1:5000/
.
- CameraVision.py: Extracts frames from videos, specifically every 21st frame.
- ComputerVision.py: Processes videos to extract text, leaning on
ImageVision.py
for detecting text inside red boxes. - ImageVision.py: Discovers red boxes in images, extracting text from these zones using EasyOCR.
- app.py: The Flask UI, showcasing extracted frames and detected text.
The web UI is built using Flask, allowing users to visualize text detection results:
- index.html: Main UI for viewing images and extracted text.
- style.css: Contains styling for the UI.
For a detailed walkthrough of the UI components, view the code in the ui
directory.
Before contributing, please review the guidelines in CONTRIBUTING.md. Adherence to these rules ensures smooth collaboration and code integration.