Li-Minn Ang’s research while affiliated with University of the Sunshine Coast and other places

Forests worldwide play a critical role in biodiversity conservation and climate regulation, yet they face unprecedented challenges, particularly from wildfires. Early wildfire detection is essential for preventing rapid spread, protecting lives, ecosystems, and economies, and mitigating climate change impacts. Traditional wildfire detection methods relying on human surveillance are limited in scope and efficiency. However, advancements in remote sensing technologies offer new opportunities for more efficient and comprehensive detection. This study highlights the integration of satellite sensors, capable of detecting thermal anomalies, smoke plumes, and vegetation health changes, with machine learning, particularly Support Vector Machines (SVMs), to enhance detection efficiency and accuracy. These algorithms analyse satellite data to identify fire patterns and provide near real-time alerts. SVMs’ adaptability over time improves performance, making them suitable for evolving fire regimes influenced by climate change. Focusing on the Wolgan Valley in Eastern Australia, the study utilised Landsat-8 imagery and SVMs to detect active fires and classify burned areas. Results demonstrated that combining various spectral bands, such as the Shortwave Infrared (SWIR) and Near-Infrared (NIR), enhances the identification of active fires and smoke. The introduction of the Normalized Difference Fire Index (NDFI) further refines detection capabilities by leveraging distinct spectral characteristics from Landsat 8 imagery. Despite the promise of these technologies, challenges such as data availability and model interpretability remain. Future research should focus on integrating diverse data sources, advancing machine learning techniques, developing real-time monitoring systems, addressing model interpretability, integrating unmanned aerial vehicles, and considering climate change impacts. This study underscores the potential of machine learning algorithms and innovative indices like NDFI to improve wildfire detection and management strategies, ultimately enhancing our ability to protect lives and ecosystems in fire-prone regions.

The emotional response of robotics is crucial for promoting the socially intelligent level of human–robot interaction (HRI). The development of machine learning has extensively stimulated research on emotional recognition for robots. Our research focuses on emotional gaits, a type of simple modality that stores a series of joint coordinates and is easy for humanoid robots to execute. However, a limited amount of research investigates emotional HRI systems based on gaits, indicating an existing gap in human emotion gait recognition and robotic emotional gait response. To address this challenge, we propose a Gait-to-Gait Emotional HRI system, emphasizing the development of an innovative emotion classification model. In our system, the humanoid robot NAO can recognize emotions from human gaits through our Trajectories-Aware and Skeleton-Graph-Aware Spatial–Temporal Transformer (TS-ST) and respond with pre-set emotional gaits that reflect the same emotion as the human presented. Our TS-ST outperforms the current state-of-the-art human-gait emotion recognition model applied to robots on the Emotion-Gait dataset.

Effective wildfire susceptibility mapping is critical for managing forest fire risks, especially in vulnerable regions like Lamington National Park, Queensland, Australia. This study compares Artificial Neural Networks (ANN) and U-Net Convolutional Neural Networks (CNN), highlighting their strengths and limitations. ANN, known for its simplicity and efficiency, is ideal for rapid assessments and resource-constrained applications, achieving higher overall accuracy. Conversely, the U-Net model excels in precision, recall, and F1 score, leveraging its advanced spatial feature extraction and localization capabilities. This makes U-Net particularly suitable for detailed mapping and critical decision-making in complex terrains or high-stakes scenarios. Utilizing geospatial datasets, including elevation, slope, vegetation indices, and proximity to infrastructure, the study demonstrates that ANN is effective for broad applications, while U-Net addresses scenarios where minimizing false positives and negatives is essential. These findings guide stakeholders in selecting the appropriate model for tailored wildfire management and proactive mitigation strategies.

Customized binary convolutional neural network (BCNN) architectures, which are implemented on hardware technologies, give significant advantages for computational efficiency and hardware acceleration. The deployment of these customized BCNNs in several real-time domains such as edge devices, embedded systems and other resource-constrained hardware platforms is becoming increasingly important. BCNN architectures, with their simplified representation of binarized weights and activations, significantly reduce computational and memory bandwidth requirements. On the one hand, the straightforward binarization of full precision CNN architectures achieves the hardware simplification. On the other hand, the binarization process suffers from the reduction in the algorithm performance. It would be highly desirable to improve the computational efficiency of the BCNN architectures through algorithmic and hardware-level optimizations without significantly affecting the algorithm performance for implementation on hardware technologies. The usage of the neural architecture search (NAS) to optimize the BCNN architectures is becoming a promising approach. This paper proposes and illustrates efficient designs and customized BCNN architectures with examples for two edge applications (compressed sensing and image super-resolution). Our designs improve the computational efficiency of the BCNN architectures through algorithmic and hardware-level optimizations without significantly affecting the algorithm performance. In some cases, the NAS-optimized BCNN architectures perform better than the full precision CNN architectures. Hardware analysis substantiates the computational effectiveness of the proposed architectures.

In recent years, there has been a growing interest in deep learning-based clustering. A recently introduced technique called DipDECK has shown effective performance on large and high-dimensional datasets. DipDECK utilises Hartigan’s dip test, a statistical test, to merge small non-viable clusters. Notably, DipDECK was the first deep learning-based clustering technique to incorporate the dip test. However, the number of initial clusters of DipDECK is overestimated and the algorithm then randomly selects the initial seeds to produce the final clusters for a dataset. Therefore, in this paper, we presented a technique called UNSDipDECK , which is an improved version of DipDECK and does not require user input for datasets with an unknown number of clusters. UNSDipDECK produces high-quality initial seeds and the initial number of clusters through a deterministic process. UNSDipDECK uses the unique closest neighbourhood and unique neighbourhood set approaches to determine high-quality initial seeds for a dataset. In our study, we compared the performance of UNSDipDECK with fifteen baseline clustering techniques, including DipDECK, using NMI and ARI metrics. The experimental results indicate that UNSDipDECK outperforms the baseline techniques, including DipDECK. Additionally, we demonstrated that the initial seed selection process significantly contributes to UNSDipDECK ’s ability to produce high-quality clusters.

The evolution of super-resolution (SR) technology has seen significant advancements through the adoption of deep learning methods. However, the deployment of such models by resource-constrained devices necessitates models that not only perform efficiently, but also conserve computational resources. Binary neural networks (BNNs) offer a promising solution by minimizing the data precision to binary levels, thus reducing the computational complexity and memory requirements. However, for BNNs, an effective architecture is essential due to their inherent limitations in representing information. Designing such architectures traditionally requires extensive computational resources and time. With the advancement in neural architecture search (NAS), differentiable NAS has emerged as an attractive solution for efficiently crafting network structures. In this paper, we introduce a novel and efficient binary network search method tailored for image super-resolution tasks. We adapt the search space specifically for super resolution to ensure it is optimally suited for the requirements of such tasks. Furthermore, we incorporate Libra Parameter Binarization (Libra-PB) to maximize information retention during forward propagation. Our experimental results demonstrate that the network structures generated by our method require only a third of the parameters, compared to conventional methods, and yet deliver comparable performance.