The implementation of novel intelligent applications has resulted in security enhancement in the way information is managed. The processing, storing, and functional aspects of computer systems have all been impacted by this revolution. Cloud computing has become a widely adopted and extensively utilized concept. Unfortunately, there are still some obstacles that prevent the widespread use of cloud computing. Edge computing refers to a decentralized kind of computing that allows the processing of data at its source, extending beyond traditional centralized systems. This research study introduces an innovative and efficient approach for identifying human faces for access control and security monitoring using sophisticated deep-learning algorithms. The proposed approach employs the results of two models trained using VGG face and ResNet architectures, respectively. Integrating these models with the deep face model enhances the accuracy and reliability of facial recognition. The process involves extracting distinguishing characteristics from facial photos using pre-trained VGG face and ResNet models. Subsequently, the deep face model is used to merge the characteristics, hence permitting improved portrayal and recognition of facial features. The proposed methodology has been verified by empirical evaluations, which have shown its effectiveness in improving precision and robustness in tasks associated with human facial recognition. The work significantly enhances the advancement of deep learning techniques in the field of face recognition applications for security enhancement.