Postgraduate Seminar by Nazrul and Michelle
Nazrul Ismail and Michelle Luon presented their Masters research project on Monocular Visual SLAM and IOT in Agriculture respectively.
On 24th November 2021, we organised two postgraduate seminar talks:
Analysis on Geometry and Deep Learning based methods for Monocular Visual SLAM by Nazrul Ismail
Abstract:
Estimating camera poses from sequences of images are the core problem formulation for visual odometry (VO) and simultaneous localization and mapping (SLAM). The classical methods involve in extracting features such as point edges and perform sampling based outlier rejection have been the common standards over the decade. The success of deep learning has motivated researchers towards an end to end learning approach. Although, results have shown that neural network are capable of learning through sequences of images, most of them have not been as accurate, robust and generalizable as the conventional methods are. In this study, we present a hybrid of both learning and geometry based methods for a Monocular SLAM pipeline consisting optical flow, depth prediction, tracking and mapping.
Speaker bio:
Nazrul Ismail received his BSc in Computer Science from Universiti Brunei Darussalam in 2019 He is now a Msc student in Universiti Brunei Darussalam(UBD) in the Department of Digital Science, Universiti Brunei Darussalam He has worked as a research assistant at UBD involving Automated plants species recognition system and had previous experience in industry as an intern at the digitalisation department, Brunei Shell Petroleum doing anomaly activity detection for Health Safety and environment His research interests include in the intersection of Machine Learning and Computer vision in interdisciplinary fields
He is currently under the supervision of Dr Owais Malik and Dr Wee Hong Ong
Automated Soil Monitoring for IOT System in Agriculture by Michelle Luon
Abstract:
Advancements in technology and the continuous internet manifestation architecture have transformed many sectors in society. The automated soil monitoring for IoT improves overall agriculture production. The study will be conducted with a low cost sensor with a low tech savvy farmers. Hence, an automated setup mechanism in agricultural IOT system which could help farmers manage their farm, with little or no technical background. The scope of this paper is to help and guide farmers/relatives with another smart technology involve in farming. Farmers can monitor the parameters from the sensors that affects the farming. To conclude, farmers can install the IoT system and able to monitor the parameters from the sensors that affects the farming.
Speaker bio:
Michelle Esher Sasha Luon received the BSc Hons in Computer Science from University of Chester (in partnership with Laksamana College Brunei ), in August 2020 She did her Final year project in AniTemb Augmented Reality for Temburong wildlife During her free time, she usually helps her family in running a family business and freelancing website designer She is now a MSc by research student in the School of Digital Science, University Brunei Darussalam Her research interest includes the field of internet of things in Agriculture.
She is currently under the supervision of Dr Daphne Teck Ching Lai, Dr Wee Hong Ong and Dr Rosyzie Anna Apong.
#seminar #postgraduate #ubddigitalscience
—
Universiti Brunei Darussalam
Jalan Tungku Link, BE1410
Negara Brunei Darussalam
office.sds@ubd.edu.bn