Number Plate Detector

This project detects and reads vehicle number plates from images using OpenCV and Tesseract OCR.

Features

• Loads an image and resizes it for processing
• Converts the image to grayscale and applies filtering
• Detects edges and contours to locate the number plate
• Extracts the number plate region and applies thresholding
• Uses Tesseract OCR to read the number plate text
• Saves intermediate and final results as images

Requirements

• Python 3.10+
• OpenCV (cv2)
• pytesseract
• matplotlib
• Tesseract OCR (installed on your system)

Installation

1. Clone this repository:

   git clone https://github.com/yourusername/number-plate-detector.git
   cd number-plate-detector

2. Install Python dependencies:

   pip install opencv-python pytesseract matplotlib

3. Install Tesseract OCR:

   • Ubuntu:
     sudo apt-get install tesseract-ocr
   • Windows:
     Download from Tesseract at UB Mannheim

4. Place your input image as fi.jpeg in the project directory.

Usage

Run the script:

python app.py

• The script will process fi.jpeg and save output images at each step.
• The detected number plate and OCR result will be saved as images.

Output Files

• output_6_1_original.jpg: Resized original image
• output_6_2_gray.jpg: Grayscale image
• output_6_2_filtered.jpg: Filtered image
• output_6_3_edges.jpg: Edge-detected image
• output_6_3_contours.jpg: Image with contours drawn
• output_6_4_plate.jpg: Cropped number plate region
• output_6_5_thresh.jpg: Thresholded plate image
• output_6_6_ocr_result.jpg: Plate image with OCR result

Notes

• Make sure the path to Tesseract in app.py matches your system installation.
• For best results, use clear images of number plates.

License

MIT License

---

Created with OpenCV and Tesseract