Sparse Autoencoder (SAE) based Image Classification with Enhanced Security for College Attendance System

An innovative attendance system that uses image fusion techniques to combine facial and iris images to improve recognition accuracy. Using a sparse autoencoder (SAE), we efficiently combine these two distinct biometric methods to create a comprehensive fused image that preserves the important features of both. This fusion solves the disadvantages of traditional recognition methods, such as sensitivity to illumination changes and face expressions, while leveraging the consistent uniqueness of iris patterns. The merged images are then processed by a Convolutional Neural Network (CNN), which excels at extracting hierarchical features for accurate classification. By combining facial and iris images, the system benefits from the complementary strengths of the two modalities. The SAE and CNN are produced with an accuracy of 92.74 % of fusion images. The proposed system not only improves overall authentication accuracy but also improves security by reducing the risk of unauthorized access through dual biometric verification