Build a PHP Camera Emotion Detector with FFmpeg and Fer2013

Cameras are ubiquitous, but recognizing human emotions from them remains challenging; advances in artificial intelligence now make webcam‑based emotion detection feasible. This article explains how to control a camera with PHP and analyze facial expressions to infer emotions.

1. Preparation

First, install the PHP GD extension for image processing and the V4L2 utilities for webcam access on a Linux system.

sudo apt-get install php7.4-gd

sudo apt-get install v4l-utils

2. Capturing the Video Stream

Use PHP's shell_exec to run an FFmpeg command that captures a single frame from /dev/video0, saves it as an image, and then display it with an <img> tag.

<?php
function getVideoStream() {
    $cmd = "ffmpeg -i /dev/video0 -vf fps=1 -s 1280x720 -f image2 -frames 1 /path/to/image.jpg";
    shell_exec($cmd);
    return "/path/to/image.jpg";
}
$videoStream = getVideoStream();
echo "<img src='$videoStream'>";
?>

3. Emotion Recognition with Fer2013

Integrate the open‑source Fer2013 model by invoking a Python script from PHP. The script receives the model file and the captured image, returns the detected emotion.

<?php
function getEmotion($imagePath) {
    $modelPath = "path/to/Fer2013/model.hdf5";
    $cmd = "python3 scripts/emotion_classification.py $modelPath $imagePath";
    $emotion = shell_exec($cmd);
    return $emotion;
}
$emotion = getEmotion($videoStream);
echo "Current emotion: $emotion";
?>

4. Complete Example

The following script combines video capture and emotion detection, outputting both the webcam image and the inferred mood.

<?php
function getVideoStream() {
    $cmd = "ffmpeg -i /dev/video0 -vf fps=1 -s 1280x720 -f image2 -frames 1 /path/to/image.jpg";
    shell_exec($cmd);
    return "/path/to/image.jpg";
}
function getEmotion($imagePath) {
    $modelPath = "path/to/Fer2013/model.hdf5";
    $cmd = "python3 scripts/emotion_classification.py $modelPath $imagePath";
    $emotion = shell_exec($cmd);
    return $emotion;
}
$videoStream = getVideoStream();
$emotion = getEmotion($videoStream);
echo "<img src='$videoStream'>";
echo "Current emotion: $emotion";
?>

Conclusion

By following these steps you can capture webcam video with PHP, process the frame, and apply the Fer2013 model to recognize facial emotions, providing a simple yet functional starting point for integrating camera‑based emotion detection into your projects.