Malpractice Detection in Online Proctoring using Deep Learning
Immadi Dhatri Raga Priya, Ponnaluri Sree Ramya, Mulugu Veera Vamshi, Bhimavarapu Chandana, M. Kameswara Rao · 2023
One of the most effective image processing applications, face recognition is crucial in the technical world. An ongoing problem with authentication is the ability to recognize a person's face, particularly when it comes to tracking students’ movements in an examination. It is a pupil identification method that employs face biostatistics based on multiple technologies. The creation of this system aims to digitally replace physical invigilation with the digital view. The manual recording makes it harder to concentrate on numerous students. HOG classifiers, CNN, SVM, Generative Adversarial Networks, and Gabor filters are used in the suggested system. Face detection systems have grown over years and the industry has paid close attention to this topic. Numerous studies have been conducted on face and eye identification. Reports will be generated and saved following face recognition. The system is tested under different conditions, which include head movements, and changing distances between the person and the cameras. Overall accuracy is determined after testing. This system is efficient to reduce malpractice in less time. The created system is inexpensive and requires little installation.