This project implements an AI-based attendance system utilizing the Local Binary Patterns Histogram (LBPH) algorithm for face recognition. LBPH is a simple yet effective algorithm for facial feature extraction, especially suitable for controlled environments, making it ideal for attendance systems.

Face recognition involves identifying or verifying a person from a facial image. It has two key tasks:

• Face Detection: Locating faces within an image.
• Face Recognition: Using the detected facial regions to recognize individual identities.

Our system uses LBPH to recognize faces for automatic attendance marking.

Local Binary Patterns Histogram (LBPH) is a texture operator that labels pixels by comparing each pixel to its neighbors, creating a binary pattern. It has four primary parameters:

• Radius: Defines the circular local binary pattern radius, typically set to 1.
• Neighbors: Number of sample points around the central pixel, usually 8.
• Grid X and Grid Y: Defines the grid dimensions, affecting the histogram size.

1. Parameter Selection: Set radius, neighbors, and grid dimensions.
2. Training: Use a dataset of labeled facial images to train the model.
3. LBP Operation: Convert each pixel’s neighborhood into a binary pattern, creating an intermediate image.
4. Histogram Extraction: Divide the image into grids, compute histograms, and concatenate them.
5. Face Recognition: Compare histograms of the input image with stored histograms using distance metrics (e.g., Euclidean distance).

LBPH is provided by the OpenCV library, supporting multiple languages, including C++ and Python, making it versatile for AI projects. You can also find an LBPH implementation in Go on GitHub.

##Video Explaination

Feel free to contribute to improving this project!