Projects Included
- Real-Time Object Detection using YOLOv5
This project demonstrates how to perform object detection on video streams using the YOLOv5 deep learning model.
Features
Uses YOLOv5 pretrained model
Detects multiple objects in video frames
Displays bounding boxes, class labels, and confidence scores
Saves processed video with detections
Technologies Used
Python
PyTorch
OpenCV
YOLOv5
Example Workflow
Load YOLOv5 model using PyTorch Hub.
Read video frames using OpenCV.
Perform object detection on each frame.
Draw bounding boxes around detected objects.
Save the processed video output.
- Hybrid Image Generation using Frequency Filters
This project generates hybrid images by combining high-frequency and low-frequency components from two different images.
Concept
Hybrid images appear different depending on viewing distance:
Close distance → high frequency image visible
Far distance → low frequency image visible
Implementation Steps
Convert images to grayscale
Apply Gaussian blur for low-pass filtering
Subtract blurred image to obtain high-pass filtered image
Combine both filtered images
Technologies Used
Python
OpenCV
NumPy
Matplotlib
- Ridge Regression Model
This project implements Ridge Regression, a regularized linear regression technique that helps prevent overfitting.
Features
Interactive input of dataset
Accepts multiple features and samples
Adjustable regularization parameter (alpha)
Outputs regression coefficients and intercept
Technologies Used
Python
NumPy
Scikit-Learn Installation
Clone the repository:
git clone https://github.com/your-username/computer-vision-ml-projects.git cd computer-vision-ml-projects
Install required dependencies:
pip install numpy opencv-python torch scikit-learn matplotlib Running the Projects Run Ridge Regression python ridge_regression.py Run YOLOv5 Object Detection python yolov5_detection.py Run Hybrid Image Generation python hybrid_image.py Repository Structure computer-vision-ml-projects │ ├── ridge_regression.py ├── yolov5_detection.py ├── hybrid_image.py ├── README.md └── sample_videos_images Learning Outcomes
Through these projects I explored:
Deep learning based object detection
Image processing using frequency filtering
Regularization techniques in machine learning models
Practical usage of OpenCV, PyTorch and Scikit-Learn
Future Improvements
Train a custom YOLO model on a specific dataset
Implement real-time webcam detection
Add model evaluation metrics
Create a simple web interface for object detection