hello its my first time doing the text on it we r going to learn about the opencv here from loading img to detecting img and much more
FaceRecog-CV is a modular Computer Vision project built to explore and implement the pipeline from basic image processing to advanced facial recognition. The project demonstrates a progression from pixel-level manipulation (thresholding, masking, histograms) to building a trainable face recognition model using OpenCV and Deep Learning concepts.
This repository contains trained artifacts (features.npy, labels.npy) capable of identifying individuals from both static images and real-time streams.
- Model Training: Custom training pipeline that extracts ROIs (Regions of Interest) and serializes features into NumPy arrays.
- Inference: Real-time face identification using pre-trained
haar_face.xmlclassifiers. - Scalability: Supports multi-face detection in group environments.
- Utilizes Haar Cascades for robust face and feature detection.
- Implements coordinate extraction to draw dynamic bounding boxes around detected subjects.
- Noise Reduction: Implementation of Gaussian, Median, and Bilateral filtering for image smoothing.
- Edge Detection: Canny edge detection algorithms for structural analysis.
- Color Analysis: Histogram computation for color distribution and channel splitting/merging.
- Geometric Transformations: Masking and bitwise operations for ROI isolation.
- Language: Python
- Computer Vision: OpenCV (
cv2) - Data Manipulation: NumPy
- Visualization: Matplotlib (
plt)
├── Section 1 - Basics/ # Image smoothing, bitwise ops, histograms
├── Section 2 - Advance/ # Haar Cascades, Face Detection algorithms
├── models/ # Serialized models (features.npy, labels.npy)
├── src/ # Source code for training and recognition
├── requirements.txt # Dependencies
└── README.md # Project documentation