https://github.com/prishabhanot/facial_recognition_pca

A face recognition system using Principal Component Analysis (PCA) for dimensionality reduction and a Support Vector Machine (SVM) classifier for classification. PCA extracts essential features (eigenfaces) from facial images, significantly reducing computational complexity while retaining critical information for accurate recognition.
https://github.com/prishabhanot/facial_recognition_pca

data eigenfaces facial-recognition pca python reducing-computational-complexity reducing-data-dimensions svm-classifier

Last synced: over 1 year ago
JSON representation

A face recognition system using Principal Component Analysis (PCA) for dimensionality reduction and a Support Vector Machine (SVM) classifier for classification. PCA extracts essential features (eigenfaces) from facial images, significantly reducing computational complexity while retaining critical information for accurate recognition.

• Host: GitHub
• URL: https://github.com/prishabhanot/facial_recognition_pca
• Owner: Prishabhanot
• Created: 2025-02-21T08:12:19.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2025-02-21T09:17:42.000Z (over 1 year ago)
• Last Synced: 2025-02-21T09:27:37.664Z (over 1 year ago)
• Topics: data, eigenfaces, facial-recognition, pca, python, reducing-computational-complexity, reducing-data-dimensions, svm-classifier
• Language: Python
• Homepage:
• Size: 2.32 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Face Recognition Using PCA and SVM

## Overview

This project implements a face recognition system using Principal Component Analysis (PCA) for dimensionality reduction and a Support Vector Machine (SVM) classifier for classification. The dataset consists of face images represented as pixel values.

## Results

Reduced feature dimensions from 4096 to 200 while retaining significant variance. Achieved a classification precision score of 97%, recall of 96%, and F1-score of 96%. PCA efficiently finds a lower-dimensional space that preserves key facial variations for classification.

![EigenFaces](Facial_Rec.png)

## Requirements

Ensure you have the following dependencies installed before running the script:

pip install numpy pandas matplotlib scikit-learn

## Dataset

The dataset (face_data.csv) contains face images in pixel form with an associated target label.

Each image is represented as a flattened array of pixel intensities (64x64 = 4096 features).

## Steps Performed

- Load and Visualize the Dataset

- Read face_data.csv into a Pandas DataFrame.

- Display the first few rows (optional).

- Extract target labels and pixel features.

- Show original grayscale face images.

- Preprocessing and Splitting Data

- Split the dataset into training (80%) and testing (20%) sets.

- Apply PCA for Dimensionality Reduction

- Fit PCA with 200 principal components.

- Plot the cumulative variance explained by each component.

- Display eigenfaces to visualize principal components.

- Train an SVM Classifier

- Transform training data using PCA.

- Train an SVM with an RBF kernel, C=1000, and gamma=0.01.

- Evaluate Model Performance

- Transform test data using PCA.

- Predict labels using the trained SVM.

- Print classification report (precision, recall, F1-score).

## Running the Script

Simply execute the Python script:

python face_recognition_pca.py