Skip to content

Latest commit

 

History

History
94 lines (70 loc) · 3.81 KB

README.md

File metadata and controls

94 lines (70 loc) · 3.81 KB
Raw

Face Recognition and Search Application

Overview

This repository contains a face recognition and search system built with Python. The application allows for the detection, encoding, and storage of facial data in a MongoDB database. It supports searching for similar faces based on the stored data and provides a graphical user interface (GUI) to manage the face search and display results.

Features

  • Face detection and recognition: Extract faces from images, encode them, and save unique faces to a database.
  • Database management: Stores face encodings and image metadata using MongoDB.
  • Face search functionality: Find similar faces from the database using face encoding comparison.
  • Graphical User Interface (GUI): Built with Tkinter for managing face data and searching faces. (to be updated)
  • Logging: Application logs all key actions and events to a log file and console.

Project Structure

.
├── database.py        # Handles MongoDB operations and face encoding storage/retrieval
├── face_detection.py  # Processes images, detects faces, and manages face data storage
├── gui.py             # GUI implementation for searching and displaying face data
├── main.py            # Entry point for running the GUI application
├── search.py          # Contains functions to search faces by ID
├── images/            # Folder where input images are stored
└── unique_faces/      # Folder where unique face images are saved

Dependencies

Make sure you have the following libraries installed before running the code:

  • pymongo: For database operations.
  • face_recognition: For facial recognition and encoding.
  • opencv-python: For image processing.
  • pillow: For handling images in the GUI.
  • tkinter: For GUI development (comes pre-installed with Python).
  • numpy: For handling face encoding data.

Install all dependencies via pip:

pip install pymongo face_recognition opencv-python pillow numpy

or

pip install -r requirements.txt

Usage

1. Setting Up MongoDB

Ensure that MongoDB is running on localhost:27017 and the database face_recognition_db is created. The face encodings and metadata will be stored in the faces_collection.

2. Running the Face Processing Script

To extract and save unique faces from images:

  1. Place your images in the images/ folder.
  2. Run the following script to detect faces and store them in MongoDB:
python face_detection.py

This will:

  • Extract faces from the images.
  • Store each unique face's encoding and associated metadata in MongoDB.
  • Save each detected face as an image in the unique_faces/ folder.

3. Running the GUI Application

To launch the face search interface:

python main.py

The GUI will display all unique faces. You can select a face and search for its occurrences in the images.

4. Searching Faces

  • Select a face: Click on a face from the GUI.
  • Search for occurrences: Click the "Search Selected Face" button to find where this face appears across the images.

Logging

All key operations (face detection, saving, searching) are logged into app.log. You can change the logging level for more or less verbosity by modifying the logging.basicConfig configuration in each file.

Customization

  • Face Similarity Tolerance: Adjust the FACE_SIMILARITY_TOLERANCE constant in database.py to change how strict the face similarity comparison is.
  • Unique Faces Directory: Change the folder where unique faces are saved by modifying the unique_faces_folder path in face_detection.py.

Troubleshooting

  • Ensure MongoDB is running and accessible at localhost:27017.
  • Make sure the required Python dependencies are installed.
  • Check app.log for detailed error logs.

License

This project is licensed under the MIT License.