Roozbeh Rajabi
Faculty Member
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (12)
Year:
2026
Year:
2025
Year:
2025
Year:
2025
The development of new green energy sources is crucial for powering the technologies and industries of the future, including next-generation transportation systems. Hydrogen is appealing as a clean and renewable energy source, but its production often generates significant CO₂ emissions. In contrast, geologic hydrogen naturally occurs in the subsurface and could be available in large quantities without additional pollution. It is formed through serpentinization, a low-temperature metamorphic process in which ultramafic rocks react with water to produce serpentinite. Hyperspectral imaging has an advantage of collecting hundreds of spectral bands and today it has many applications including environmental analysis, mineral identification, industrial quality assessment, etc. Hyperspectral imaging techniques including spaceborne, airborne, ground, and underwater systems can play a vital role in natural hydrogen exploration. In this presentation, we review three major applications of imaging spectroscopy for geologic hydrogen exploration: (1) the identification of key minerals involved in the serpentinization process, such as olivine and serpentine; (2) the detection of hydrogen seeps through vegetation indices monitoring around fairy circles; and (3) spectral analysis of drill core samples for subsurface insights. Empirical studies have reported promising results across multiple case studies worldwide.
Year:
2025
Identifying critical minerals is vital for national security, economic stability, and technological advancement. These minerals are essential across industries such as aerospace, high-tech electronics, and renewable energy technologies, including wind turbines and electric vehicle batteries, which depend on rare earth elements like cobalt, lithium, graphite, gallium, and tantalum. Ensuring a stable and secure supply of these resources is a strategic priority, as disruptions from geopolitical tensions, trade restrictions, or resource scarcity can significantly impact key sectors. Hyperspectral imaging systems are powerful tools for mineral identification, providing detailed spectral information across a wide range of wavelengths. However, the relatively low spatial resolution of airborne data often results in spectral mixing, where signals from multiple materials, including different minerals and vegetation cover, are combined within a single pixel. In this presentation, we evaluate the effectiveness of spectral unmixing techniques to generate more accurate abundance fraction maps of various minerals by separating the vegetation spectra. The study area is in Coal Creek, Alaska. Hyperspectral data is acquired using airborne VNIR/SWIR camera system. The data were processed using PARGE and ATCOR software for georeferencing and atmospheric correction. Mineral mapping was performed using the USGS Mineral Identification and Characterization Algorithm (MICA). To further analyze and address the effect of vegetation on the mineral maps, autoencoder-based spectral unmixing methods were applied. This technique allowed for the separation of vegetation and mineral fractions, enhancing the reliability and interpretability of the resulting mineral abundance maps.
Year:
2025
Nonintrusive load monitoring (NILM) is an important technique for energy management and conservation. In this paper, a deep learning model based on an attention mechanism, temporal pooling, residual connections, and transformers is proposed. This article presents a novel approach for NILM to accurately discern energy consumption patterns of individual household appliances. The proposed method entails a sequence of layers, including encoders, transformers, attention, temporal pooling, and residual connections, offering a comprehensive solution for NILM while effectively capturing appliance-specific energy usage in a household. The proposed model was evaluated using UK-DALE, REDD, and REFIT datasets in both seen and unseen cases. It shows that the proposed model in this paper performs better than other methods stated in other papers in terms of F1-score and total error of the results (in terms of SAE). This model achieved an F1-score equal to 92.96 as well as a total SAE equal to −0.036, which shows its effectiveness in accurately diagnosing and estimating the energy consumption of individual home appliances. The findings of this research show that the proposed model can be a tool for energy management in residential and commercial buildings.
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2023
Year:
2023
Collaborators (2)
ABOUZAR ESTEBSARI
Associate Professor
London South Bank University
Hossein Ahmadi
Politecnico di Torino
Make your next move.
Prepare your application
Review drafts before you send them.