---
title: "Face Recognition using OpenCV"
url: "https://ridwaanhall.com/projects/face-recognition-using-opencv"
markdown: "https://ridwaanhall.com/projects/face-recognition-using-opencv.md"
summary: "A real-time face recognition system in Python and OpenCV, with Haar Cascade detection and LBPH recognition."
---

# Face Recognition using OpenCV

> A real-time face recognition system in Python and OpenCV, with Haar Cascade detection and LBPH recognition.

- Category: computer-vision
- Status: Completed
- Started: March 5, 2024
- Built with: OpenCV, Python, NumPy
- Source: https://github.com/ridwaanhall/face-recognition-using-OpenCV

## About

### The problem

Face recognition is often presented as a black box, but the classic pipeline of detection, training and recognition can be built and understood with standard tools.

### The approach

I built a complete real-time system in Python and OpenCV: Haar Cascade classifiers detect faces and eyes, and the LBPH algorithm recognises them.

- Scripts to collect and train face images for multiple users with unique IDs.
- Real-time recognition on the live camera feed, with optimised grayscale processing.
- Configurable camera index, images per user and cascade files.

### The result

The system recognises multiple users live and serves as a clear, working reference for the classical computer vision approach to faces.

## Features

- **Face Detection**: Uses Haar Cascade Classifiers for accurate face and eye detection.
- **Face Recognition**: Implements LBPH algorithm for robust face recognition.
- **Real-time Processing**: Processes live camera feed for instant face recognition.
- **Multi-user Support**: Supports multiple users with unique IDs and training data.
- **Easy Training Interface**: Simple scripts for collecting and training face images.

## Tags

face-recognition, lbph, Lbph Face Recognizer, opencv

## Elsewhere

- This page for people: https://ridwaanhall.com/projects/face-recognition-using-opencv
- Index of every page as Markdown: [llms.txt](https://ridwaanhall.com/llms.txt)
