Strengthening Public Exam Integrity: Reliable Attendance Verification and Impersonation Detection using Image Recognition Fusion of Convolutional Neural Network and K – Nearest Neighbors

The study of Detection of facial impersonation in public examinations. using Convolutional Neural Network Algorithm in Comparison with the K – Nearest Neighbors Algorithm to Improve Accuracy. This study utilized a diverse dataset of Face images to explore detection, employing CNN and KNN algorithms. The CNN model, a deep learning approach, was trained for feature extraction and classification, while the KNN algorithm served as a classical machine learning comparator. Comprehensive performance evaluation metrics, including sensitivity, accuracy, specificity were employed. Ethical data practices were maintained. The statistical analysis of results aims to ascertain the superior algorithm for Detection of facial impersonation in public exams, potentially advancing the application focus on improving security verification in public exams. A study comparing the Performance of the CNN and KNN algorithms for Detection of facial impersonation in Public Exam, the CNN algorithm demonstrated significantly higher accuracy, achieving an accuracy rate of 96.75%, whereas the KNN algorithm yielded an accuracy of 92.66%. These findings highlight the superior accuracy of the CNN model and its potential to enhance the precision of Face Detection in Public Exam compared to the KNN algorithm. The Convolutional Neural Network algorithm outperformed K – Nearest Neighbors, showcasing its potential to improve the accuracy of Detection of facial impersonation in public examinations.