Face-Detection

A small Python project that demonstrates two simple face-related applications using OpenCV:

• facedetection.py — Detects faces from the webcam and shows the number of faces detected in the video feed.
• facetracking.py — Detects a face and maps its position to two Arduino-controlled servos so a camera or mechanism can track the face.

This README explains how to install the required packages, run each script, and troubleshoot common issues.

Features

• Real-time face detection using OpenCV's Haar cascade classifier.
• Visual overlays (bounding boxes, target marker, position text).
• Optional Arduino servo control (via pyfirmata) to physically track the detected face.

Prerequisites

• Python 3.8+ (3.10/3.11 recommended)
• A webcam (built-in or USB)
• If you want servo tracking: an Arduino board, two servos, and a USB connection to a Windows COM port.

Python packages

Install the required Python packages. From PowerShell you can run:

python -m pip install opencv-python numpy cvzone pyfirmata

Notes:

• cvzone bundles some useful helpers but is optional for these scripts — opencv-python and numpy are essential.
• If you run into binary wheel errors for opencv-python, try upgrading pip first: python -m pip install --upgrade pip.

Files in this repo

• facedetection.py — Simple face detection and on-screen face count.
• facetracking.py — Face tracking with servo control (requires Arduino + pyfirmata).
• haarcascade_frontalface_default.xml — Haar cascade model used by both scripts (must be in the same folder).

Usage

Open PowerShell in the repository folder (e.g., D:\Face-Detection) and run one of the scripts.

Run the face detection demo:

python facedetection.py

Press the q key in the video window to quit.

Run the face tracking demo (hardware required):

1. Connect your Arduino to the PC and confirm the COM port (Windows Device Manager). Update port = "COM7" in facetracking.py to match your COM port.
2. Connect two servos to the Arduino pins you configured (the script uses digital pins 9 and 10 by default).
3. Start the script:

python facetracking.py

Press q in the video window to quit. The script will continuously write servo angles to the Arduino using pyfirmata.

Hardware wiring notes (high level)

• Use a common ground between the Arduino and the servos.
• Power servos from a suitable power supply if they draw more current than the Arduino's regulator can safely provide.
• Typical wiring: servo signal wires to Arduino D9 and D10, servo VCC to external 5V (or Arduino 5V if current is low), servo GND to Arduino GND.

Troubleshooting

• "Camera couldn't Access" or "The camera cannot be accessed." — Ensure no other program is using the webcam and that the correct device index (0) is valid for your system.
• Cascade file not found — Make sure haarcascade_frontalface_default.xml is in the same folder as the scripts.
• Arduino pyfirmata errors — Verify the COM port string and that the Arduino is running the StandardFirmata sketch. Install Firmata on the Arduino using the Arduino IDE: open Examples → Firmata → StandardFirmata and upload it.

Development notes

• The scripts use OpenCV's Haar cascade classifier which is fast but less accurate than modern deep-learning based detectors; it's good for simple demos and low-latency use.
• cvzone is optional; you can remove it and rely directly on opencv-python + numpy.