Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal JOIV : International Journal on Informatics Visualization Emerging Science Journal
Pang, Ying Han
Unknown Affiliation
Computer Science & IT Environmental Science

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Enhancing Land Management through U-Net Deep Learning: A Case Study on Climate-Related Land Degradation in Berembun Forest Reserve in Malaysia Chew, Yee Jian; Ooi, Shih Yin; Mohd-Razali, Sheriza; Pang, Ying Han; You Lim, Zheng
JOIV : International Journal on Informatics Visualization Vol 8, No 4 (2024)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.8.4.2948

Abstract

In the face of accelerating climate change, effective management of land resources needs innovative technological approaches. This study, conducted in the Berembun Forest Reserve, Jelebu, Malaysia, leverages advancements in geospatial technology and machine learning to address the pressing issue of land degradation, focusing on forested areas vulnerable to landslides. Utilizing high-resolution Unmanned Aerial Vehicle (UAV) imagery, the U-Net convolutional neural network model is employed for the precise classification and early detection of landslide-induced land degradation. Through a systematic analysis of 15 high-quality UAV images of 5472 x 3647 pixels, segmented into 256 x 256-pixel patches, the U-Net model demonstrated remarkable accuracy, achieving a mean Intersection-over-Union (IoU) of 0.9466. These findings underscore the model's potential to significantly enhance land management practices by providing timely and cost-effective landslide detection. Adopting such deep learning techniques is a pivotal shift towards more sustainable and resilient land management strategies in the era of climate change. This research showcases the practical application of machine learning in environmental monitoring and paves the way for future innovations. Implications for further research include integrating additional spectral bands, addressing environmental variability, and expanding applications across diverse landscapes to improve environmental monitoring, conservation efforts, and resilience strategies. Developing automated frameworks for real-time data processing and model deployment could further revolutionize the field, enabling more responsive and efficient land management practices.
HSTCN-NuSVC: A Homogeneous Stacked Deep Ensemble Learner for Classifying Human Actions Using Smartphones Raja Sekaran, Sarmela; Pang, Ying Han; Shih Yin, Ooi; Zheng You, Lim
Emerging Science Journal Vol 9, No 1 (2025): February
Publisher : Ital Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28991/ESJ-2025-09-01-026

Abstract

Smartphone-based human activity recognition (HAR) is an important research area due to its wide-ranging applications in health, security, gaming, etc. Existing HAR models face challenges such as tedious manual feature extraction/selection techniques, limited model generalisation, high computational cost, and inability to retain longer-term dependencies. This work aims to overcome the issues by proposing a lightweight, homogenous stacked deep ensemble model, termed Homogenous Stacking Temporal Convolutional Network with Nu-Support Vector Classifier (HSTCN-NuSVC), for activity classification. In this model, multiple enhanced TCN networks with diverse architectures are organised parallelly to capture hierarchical spatial-temporal patterns from raw inertial signals. Each base model (i.e., TCN) incorporates dilations and residual connections to preserve longer effective histories, allowing the model to retain longer-term dependencies. Additionally, dilations can diminish the number of trainable parameters, reducing the model complexity and computational cost. The base models’ predictions are concatenated and fed into a meta-learner (i.e., Nu-SVC) for final classification. The proposed HSTCN-NuSVC is evaluated using a publicly available database, i.e., UCI HAR, and a subject-independent protocol is implemented. The empirical results demonstrate that HSTCN-NuSVC achieves 97.25% accuracy with only 0.51 million parameters. The results exhibit the model’s effectiveness in enhancing generalisation across individuals with better accuracy and computational efficiency. Doi: 10.28991/ESJ-2025-09-01-026 Full Text: PDF
Co-Authors Chew, Yee Jian Mohd-Razali, Sheriza Ooi, Shih Yin Raja Sekaran, Sarmela Shih Yin, Ooi You Lim, Zheng Zheng You, Lim