Enhanced CNN-Based Facial Recognition Optimization in MATLAB

Facial recognition technology has become increasingly pivotal in modern security and authentication systems. This project focuses on optimizing facial recognition using an enhanced Convolutional Neural Network (CNN) architecture implemented in MATLAB. The process begins with dataset acquisition from Kaggle, comprising diverse facial images for recognition tasks. Input images undergo preprocessing, including resizing to a uniform dimension to ensure consistency and efficiency during CNN training. The system performs two primary tasks: face identification and age classification. For facial recognition, the CNN architecture classifies the unique identity (ID) of individuals from the dataset. For age recognition, the same architecture is adapted to categorize individuals into defined age groups or ranges (e.g., 5-10 years). The CNN is optimized for accuracy and efficiency, making it suitable for real-time applications. The model's performance is rigorously evaluated on a predefined sample of images to ensure its robustness and reliability. This dual-purpose system demonstrates the adaptability of CNN architectures for diverse facial analysis tasks, with promising results in terms of accuracy and operational efficiency, offering significant potential for integration into real-world applications such as access control, surveillance, and demographic analysis.

Keywords: Face Dataset, Image Processing Techniques, Deep Learning Techniques, Convolution Neural Network, Classification, Accuracy.