International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 11 Issue: 05 | May 2024
www.irjet.net
p-ISSN: 2395-0072
IMPLEMENTATION OF INTELLIGENT EXAM INVIGILATION SYSTEMUSING DEEP LEARNING ALGORITHM Megha S1, Navyashree K R2, Palabandla Vandana3, Rakshitha G4 1
Student, Electronics and Communication Engineering, Atria Institute of Technology Student, Electronics and Communication Engineering, Atria Institute of Technology 3 Student, Electronics and Communication Engineering, Atria Institute of Technology 4 Student, Electronics and Communication Engineering, Atria Institute of Technology ---------------------------------------------------------------------***--------------------------------------------------------------------2
Abstract - Educational institutions rely on exams to assess students' abilities, yet cheating persists. A computer vision- based method is proposed, utilizing CCTV to detect anomalies during physical exams. Employing You Only Look Once with residual networks, the system achieves 90% accuracy in identifying cheating behaviors. This innovative approach enhances exam integrity, employing machine learning and AI to monitor exam halls effectively. The proposed method uses You Only Look Once with residual networks as the backbone architecture to inspect cheating in exams. The obtained results show the credibility and efficiency of the proposed method. The experimental results are promising and demonstrate the invigilation of the students in the examination. In this work, achieve 90%accuracy for the detection of cheating in the classroom environment. Thisresearch introduces an innovative approach to bolster the integrity of examinations by developing machine learning andartificial intelligence for the detection of suspicious activities in examination halls.
beyond conventional surveillance methods. By analyzing real-time video feeds from examination halls, the system employs computer vision techniques and sophisticated algorithms to recognize patterns associated with potential misconduct. The research addresses the limitations of existing systems by incorporating continuous learning mechanisms, ensuring the adaptability and effectiveness of the proposed solution over time.
2. LITRATURE REVIEW Intelligent Exam invigilation system is a computer- based system that is used to monitor the students and detect the suspicious activities during examination among the students. It is designed to reduce the incidence of academic dishonesty and fair conduction of the exams by monitoring the students’ actions in real-time using a combined versions of computer vision and machine learning algorithms. The purpose of this literature review is to examine the existing research on automated invigilation system for detection of suspicious activities during the examination.
Key Words: Video surveillance, Deep learning, Data Preprocessing, segmentation, feature extraction, suspicious activities, CNN.
1. “Automated invigilation system for detection of suspicious activities duringexamination”. IRJET Published 2023
1. INTRODUCTION In the context of academic assessments, ensuring the integrity of examinations is crucial to maintaining the credibility of educational systems. With the widespread integration of technology in examination processes, there arises a pressing need for intelligent solutions that can effectively detect and prevent suspicious activities in examination halls. Traditional methods of invigilation often fall short in addressing the dynamic nature of cheating tactics, prompting the exploration of innovative approaches rooted in machine learning (ML) and artificial intelligence (AI). This study aims to contribute to the enhancement of examination security by proposing a novel system that utilizes advanced technologies to autonomously identify and flag suspicious behaviors during examinations. The integration of ML and AI allows for a more proactive and adaptive approach to examination monitoring, moving
© 2024, IRJET
|
Impact Factor value: 8.226
This system detects the suspicious activities and cheating work done by students during the examination. Detection of cheating activities in classroom is implemented by the system using YOLO (you only look once) algorithm. This model is able to process the real time automated videos using the existing data set of the students and various activities can be analyzed which is happening in the examination. And this model is designed using RCNN and with training accuracy of 99.5% and testing accuracy of 98.5% ,95% accuracy in face recognition. It can track more than 100 students and requires less computing time to get the required result than previous models. The development of quicker invigilation system can further enhance the suggested invigilation system.
|
ISO 9001:2008 Certified Journal
|
Page 602