This repository is a comprehensive guide to Computer Vision techniques, covering both theoretical insights and practical implementations. It includes hands-on exercises for image processing, feature extraction, object detection, segmentation, and deep learning-based models.
- Grayscale Conversion: Simplifies images by removing color, focusing on intensity.
- Brightness & Contrast Adjustments: Modify pixel values to enhance image visibility.
- Arithmetic & Logical Operations:
- Addition & Subtraction: Blends or highlights differences between images.
- Multiplication: Enhances contrast.
- Bitwise Operations (AND, OR, XOR, NOT): Used for masking, merging, and extracting image regions.
- Flipping & Rotations:
- Horizontal, Vertical, and Both-axis flips.
- Rotations by 90°, 180°, and 270° for orientation changes.
- Image Concatenation: Combining multiple processed images horizontally or vertically.
- Concept: Extracts edge and gradient information to detect objects efficiently.
- How It Works:
- Computes gradients in an image.
- Divides image into cells and calculates histograms of gradient directions.
- Normalizes across blocks for better accuracy.
- Used in object detection (e.g., pedestrian detection) and ML classifiers.
- Capturing Images & Videos via Webcam: Using OpenCV for real-time image acquisition.
- Reading & Displaying Videos: Frame-by-frame video processing.
- Saving Captured Images: Storing frames as images for further processing.
- Feature-based techniques (HOG, SIFT, SURF)
- Deep Learning models (CNNs, YOLO, Faster R-CNN)
- Segmentation techniques (Thresholding, Watershed, Mask R-CNN)
This project is licensed under the MIT License. Feel free to use and modify it as needed.
Contributions are welcome! You can:
- Open an issue for suggestions.
- Submit a pull request with improvements.
For any inquiries, contact Bushra Shahbaz via:
📩 Email: bsdsf21m020@pucit.edu.pk
📌 GitHub: Bushra-Butt-17