In the past, there has been tremendous work done on face detection. In this paper, we have aimed to propose a very accurate face detection system using deep learning technologies. Our designed system involves convolution neural networks which in turn learn features from facial images. The dataset that we used here consisted of different facial images consisting of different facial expressions and different lighting conditions. In this project, we have used MTCNN for face detection and kernel blur analysis, then apply Lucy–Richardson deconvolution and CNN-based feature extraction for enhancing facial recognition in blurry CCTV images. Our system is trained with the help of a large-scale dataset and identification of masks as well. Additionally, our project also works on accuracy and robustness. Experimental results demonstrate the effectiveness of the proposed approach in detecting faces with high accuracy while maintaining computational efficiency, making it suitable for real-time applications. In this paper, we will elaborate that this system has the potential to contribute to various domains including security, surveillance, and human computer interaction.