The Brunel Innovation Centre (BIC) is a world class research and technology centre that sits between the knowledge base and industry offering high quality research in an innovative environment focused on non-destructive testing, condition and structural health monitoring, power ultrasonics and allied technologies covering a range of materials, sensors, electronics and software systems supporting partners in industry to transfer academic research into industrial application. BIC pursues initiatives that span national and international platforms including Innovate UK, EPSRC and EC. The Centre has been building a strong portfolio of projects in line with its multinational interdisciplinary vision. http://bura.brunel.ac.uk/handle/2438/25441 2026-04-19T13:00:05Z 2026-04-19T13:00:05Z Ayodeji, A Teyeb, A Abbas, MAA Bass, P Bass, E Bandara, PD Jayasinghe, UK Griffiths, J El Masri, E http://bura.brunel.ac.uk/handle/2438/32958 2026-03-11T03:01:47Z 2025-09-10T00:00:00Z

Title: Language-guided zero-shot segmentation with multi-angle reprojection for point cloud analysis Authors: Ayodeji, A; Teyeb, A; Abbas, MAA; Bass, P; Bass, E; Bandara, PD; Jayasinghe, UK; Griffiths, J; El Masri, E Abstract: Virtual Reality applications increasingly demand accurate 3D representations of real-world environments. While LiDAR point clouds capture physical spaces with high fidelity, they typically lack semantic labels, limiting their direct use for tasks such as object recognition, interaction modeling, and automation in immersive environments or digital twin systems. We present a LAnguage-guided zero-shot 3D SEgmentation and Reprojection tool (LASER), an engineered zero-shot segmentation tool that extends the state of the art by introducing language-guided 3D object detection for enhanced usability and accuracy. Unlike its predecessors, LASER uses an ensemble of GroundingDINO and Segment Anything Model as its backbone to process natural language queries and user-specified object categories, automated multi-view orthophoto generation with dynamic angles for optimal view selection, a confidence-weighted fusion algorithm for efficient 2D-3D reprojection, and a semantically labelled mesh output. The LASER pipeline begins by collecting point cloud data using LiDAR sensors, filtering the point cloud into ground and non-ground components, improving segmentation efficiency. It then generates multi-angle 2D orthophotos and perspective views, incorporating a user-guided angle selection module to optimise scene coverage. Then GroundingDINO detects objects based on textual descriptions, and Segment Anything Model subsequently refines these into segmentation masks. The core innovation of LASER lies in its confidence-weighted reprojection algorithm, which fuses multiple 2D segmentation results back into 3D space, ensuring higher segmentation accuracy and spatial consistency. The resulting semantically labelled assets can be exported in standard formats or iteratively refined through viewpoint adjustments or text prompt modifications. Our application of LASER to real-world 3D scans of construction sites demonstrates its effectiveness in delivering high segmentation precision, enhanced user interactivity, and seamless integration into virtual reality workflows. To comprehensively evaluate the proposed tool on diverse point cloud scans, we also presented the performance on four different test cases using two different scans (3DSES and Toronto3D) with both indoor and outdoor scenes. The results show consistent performance across scans. Finally, feature-based comparison with state-of-the-art approaches shows that LASER is an optimised tool for enriching static, open-world 3D scans with semantic labels, offering an alternative to existing state-of-the-art methods for niche applications. Description: Highlight: • Language-guided zero-shot 3D segmentation and object extraction. • User-guided multi-angle orthophoto generation for improved scene coverage. • GroundingDINO-based text-prompted detection for intuitive object detection. • Confidence-weighted fusion ensuring accurate and consistent 3D reprojection. • Flexible viewpoint selection and iterative refinement for enhanced usability.; Data availability: Data will be made available on request.

2025-09-10T00:00:00Z Ramagiri, S Zonuzi, A Lowe, S El Masri, E Teyeb, A Gan, T-H http://bura.brunel.ac.uk/handle/2438/30619 2025-01-31T03:02:12Z 2023-03-13T00:00:00Z

Title: Machine Learning and Optimisation to Improve Energy Utilisation Authors: Ramagiri, S; Zonuzi, A; Lowe, S; El Masri, E; Teyeb, A; Gan, T-H Abstract: The world is moving towards a conservative approach to fulfilling its energy needs due to inevitable uncertainty and disruptions in the supply chain. In addition, climate change, the availability of materials, and making them sustainable through recycling are other topics of high interest. Energy is a common item among all the industries, and demand for it keeps increasing due to developmental activities. In this work, we aim to improve the efficiency of utilising the available energy in the material processing industries. Mining the ore, extracting the material of interest, melting the material, and manufacturing the required components are typical processes in these industries. The manufacturing of the components also includes a heat treatment process. For example, the heat treatment process demands 20% of the total energy in a non-ferrous foundry. Pre-heating and heat treatment operations consume a significant amount of energy in the ferrous-based industry. We intend to investigate the processes in these industries and create a machine-learning model of the processes involved. Later, we use the machine learning models to build an optimization framework that provides the optimal process operating parameters to achieve the best output while using the least amount of energy. Description: Slides presented at the conference are available online at: https://www.iaria.org/conferences2023/filesENERGY23/30043_ENERGY.pdf .

2023-03-13T00:00:00Z Cosoli, G Salerno, G Calcagni, MT Pandarese, G Violini, L El Masri, E de Melo Ribeiro, H Abbas, MAA Revel, GM http://bura.brunel.ac.uk/handle/2438/30618 2025-01-31T03:02:09Z 2024-06-12T00:00:00Z

Title: How to Valorize Construction and Demolition Wastes? Beyond the State of the art Through Vision Systems and Artificial Intelligence Tools Authors: Cosoli, G; Salerno, G; Calcagni, MT; Pandarese, G; Violini, L; El Masri, E; de Melo Ribeiro, H; Abbas, MAA; Revel, GM Abstract: The efficient tracking of Construction and Demolition Wastes (CDWs) is pivotal in a perspective of sustainability and circularity in the construction sector. Sensing and digital technologies can undoubtedly play a relevant role in this context. This paper proposes an innovative approach for detection, quantification, and characterization of CDWs in order to provide information exploitable through optimized valorization routes made available via dedicated service platforms. The preliminary results are promising, and the solution will be iteratively refined and improved thanks to continuous data collection from real-world scenarios.

2024-06-12T00:00:00Z Ayodeji, A El Masri, E Williamson, T Asgar Abbas, MA Gan, T-H http://bura.brunel.ac.uk/handle/2438/30352 2024-12-18T03:01:46Z 2024-12-05T00:00:00Z

Title: State-enhanced attention network for optimisation of energy and yield in gas atomised metal powder production Authors: Ayodeji, A; El Masri, E; Williamson, T; Asgar Abbas, MA; Gan, T-H Abstract: Gas atomisation is a widely used technique for producing spherical metal powder feedstock for additive manufacturing. However, the process parameters suffer from variability and inefficiency in balancing powder yield, energy consumption, and particle size distribution. Optimising these complex, interdependent parameters pose a significant challenge. This work proposes a novel State-Enhanced Attention Network architecture in a framework that simultaneously optimises yield and energy consumption during nitrogen gas atomisation for sustainable metal powder production. The novelty lies in integrating processed long-term memory states with the attention mechanism, enabling nuanced attention weighting. This allows the model to leverage global sequence context and recent state information for improved yield and energy predictions. The proposed network is trained and integrated into a non-dominated sorting genetic algorithm to enable multi-objective optimisation. This framework evolves a set of Pareto-optimal solutions that balance trade-offs between maximising yield and minimising energy consumption. The approach is evaluated using augmented real-world data from an industrial gas atomisation plant. The proposed model demonstrates significantly improved predictive accuracy on real-world datasets, compared with baseline deep learning models. Results highlight the capabilities of the proposed technique for automated, data-driven optimisation of gas atomisation, simultaneously improving yield, energy efficiency, quality control, and sustainability. The integrated deep learning and evolutionary optimisation framework also provides an innovative solution for enhanced control of additive manufacturing powder production processes.

2024-12-05T00:00:00Z