Connie, Tee
Unknown Affiliation

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

Abnormal behavior recognition using SRU with attention mechanism Tay, Nian Chi; Connie, Tee; Ong, Thian Song; Teoh, Andrew Beng Jin; Teh, Pin Shen
International Journal of Advances in Intelligent Informatics Vol 10, No 2 (2024): May 2024
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v10i2.1385

Abstract

In response to the critical need for enhanced public safety measures, this study introduces an advanced intelligent surveillance system designed to autonomously detect abnormal behaviors within public spaces. Leveraging the computational efficiency and accuracy of a Simple Recurrent Unit (SRU) integrated with an attention mechanism, this research delivers a novel approach towards understanding and interpreting human interactions in real-time video footage. Distinctively, the model specializes in identifying two primary categories of abnormal behavior: aggressive two-person interactions such as physical confrontations and collective crowd dynamics, characterized by sudden dispersal patterns indicative of distress or danger. The incorporation of Attention mechanism precisely targets critical elements of behavior, thereby enhancing the model's focus and interpretative clarity. Empirical validation across five benchmark datasets reveals that our model not only outperforms traditional Long Short-Term Memory (LSTM) frameworks in terms of speed by a factor of 1.5 but also demonstrates superior accuracy in abnormal behavior recognition. These findings not only underscore the model's potential in preempting potential safety threats but also mark a significant advancement in the application of deep learning technologies for public security infrastructures. This research contributes to the broader discourse on public safety, offering actionable insights and robust technological solutions to enhance surveillance efficacy and response mechanisms in critical public domains.
Cheating Detection for Online Examination Using Clustering Based Approach Ong, Seng Zi; Connie, Tee; Goh, Michael Kah Ong
JOIV : International Journal on Informatics Visualization Vol 7, No 3-2 (2023): Empowering the Future: The Role of Information Technology in Building Resilien
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30630/joiv.7.3-2.2327

Abstract

Online exams have become increasingly popular due to their convenience in eliminating the need for physical exams and allowing students to take exams from remote locations. However, one of the drawbacks of online exams is that they make cheating easier, and it can be difficult for online proctoring to detect subtle movements by the students. This could lead to doubts about students' exam results' value and overall credibility. To address this pressing issue, we present a cheating detection method using a CCTV camera to monitor students' faces, eyes, and devices to determine whether they cheat during exams. If suspicious behavior indicative of cheating is detected, a warning is raised to alert the students. A custom dataset was developed to train the model. The dataset consisted of recordings of pre-determined cheating behavior by 50 participants. These videos captured various poses and behaviors encoded and analyzed using a clustering approach. The encoded clustering method continuously tracks the students' faces, eyes, and body gestures throughout an exam. Experimental results show that the proposed approach effectively detects cheating behavior with a favorable accuracy of 83%. The proposed method offers a promising solution to the growing concern about cheating in online exams. This approach can significantly enhance the integrity and reliability of online assessment processes, fostering trust among educational institutions and stakeholders.
A Robust License Plate Detection System Using Smart Device Bin Mohamad Azhar, Muhammad Darwish; Goh, Kah Ong Michael; Check Yee, Law; Connie, Tee
JOIV : International Journal on Informatics Visualization Vol 8, No 2 (2024)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.8.2.2287

Abstract

The license plate recognition (LPR) system is widely employed in various applications. However, most research studies have used a fixed camera rather than a moving one. This is because the location of the vehicle plate is nearly static and easily estimated, making the use of a static camera simple for locating and detecting the scanned license plate. Images obtained with a moving camera are highly complex due to frequent background changes. Additionally, a challenge with car plates in Malaysia is their non-standardized nature. Car owners are permitted to use any font type for their license plate number, rendering existing license plate recognition systems from other countries incapable of effectively detecting license plates on Malaysian car plates. A traditional LPR system typically requires a high-quality camera and a powerful computer for costly and bulky processing. Nowadays, many smartphones come equipped with powerful processors and cameras. Android smartphones include various libraries for modifying hardware configurations such as the camera. This paper presents a robust method for detecting Malaysia's license plate number using a convolutional neural network (CNN). The CNN model from the pre-training process is imported to the Android device and tested in real-time in an on-road driving environment, resulting in an average recognition rate of 89.37%. A comprehensive Character Recognition Analysis is also presented to demonstrate the accuracy of each character. However, there is still room for improvement in recognizing the character Q.
Traffic Violation Detection Using Computer Vision Techniques Ong, Chin Sin; Connie, Tee; Ong Goh, Michael Kah
JOIV : International Journal on Informatics Visualization Vol 8, No 3-2 (2024): IT for Global Goals: Building a Sustainable Tomorrow
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.8.3-2.2941

Abstract

The increasing number of road accidents is still a global concern.  Traditional approaches to detecting traffic violators on the road, such as radar guns and sensors, are expensive and time-consuming to maintain and install. This often results in inefficient and ineffective detection of traffic violators. This paper proposes a more cost-effective and efficient approach to traffic violation detection utilizing visual data from CCTV footage. Specifically, the method targets two common violations: crossing red lights and overtaking on double lines. In this study, YOLO is integrated for road object detection, providing the detection of vehicles and traffic lights on the road for our system. Then, the Deep SORT tracker tracks detected vehicles, ensuring continuous monitoring over time. An automated lane detection technique is formulated to identify the stopping line/lane for red light violation detection, enabling precise detection of vehicles that cross the stop lane during red light. For overtaking detection, the system detects the double line to serve as the boundary that vehicles should not cross, identifying illegal overtaking. Furthermore, point-line distance calculation is utilized to detect traffic violators by analyzing their tracked trajectories and positions. The proposed solution is evaluated using real-world CCTV footage from online repositories to reflect the real-world scenarios as closely as possible. Experimental results show that the proposed techniques achieve promising detection of real-time traffic violators, which leads to a safer environment for road users.