The Transformer model encounters challenges with variable-length input sequences, leading to efficiency and scalability concerns. To overcome this, we propose a pyramid-based hierarchical Spatial-Spectral Transformer (PyFormer). This innovative approach organizes input data hierarchically into pyramid segments, each representing distinct abstraction levels, thereby enhancing processing efficiency. At each level, a dedicated Transformer encoder is applied, effectively capturing both local and global context. Integration of outputs from different levels culminates in the final input representation. In short, the Pyramid excels at capturing spatial features and local patterns, while the Transformer effectively models spatial-spectral correlations and long-range dependencies. Experimental results underscore the superiority of the proposed method over state-of-the-art (SOTA) approaches, achieving overall accuracies of 96.28% for the Pavia University dataset and 97.36% for the University of Houston dataset. Additionally, the incorporation of disjoint samples augments robustness and reliability, thereby highlighting the potential of PyFormer in advancing HSIC. The source code is available at https://github.com/mahmad00/PyFormer .

The assessment of the urban environment is a crucial area of interest for the international community, particularly for the Kingdom of Saudi Arabia. Achieving environmental sustainability is among the goals of Saudi Vision 2030. Riyadh has recently witnessed urban expansion due to population growth, which can cause negative environmental effects unless this growth is accompanied by the development of appropriate environmental regulations and controls. Therefore, this study aimed to assess the city of Riyadh’s urban environment and its relationship to urban expansion over time. The Thermal Environmental Index (TEI), consisting of five indicators (vegetation index, urban index, moisture index, dry index, and temperature index) was constructed utilizing geospatial techniques. Pearson’s correlation analysis was performed to determine the relationship between urban expansion and urban environmental conditions. The main results show that the size of Riyadh’s urban clusters doubled by 2022, increasing by 106% compared to in 2000. The results also show that, in 2000, 2013, and 2022, areas of poor environmental conditions were mainly distributed in urban areas, whereas areas with excellent and good environmental conditions were located outside of urban areas in the northern and western parts of Riyadh. Areas with good environmental conditions continued to decline, whereas areas poor environmental conditions increased from 4%, or 240 km², in 2000 to 11% in 2022, reaching 658 km². The correlation analysis clearly showed a very strong relationship between urban expansion and the deterioration in urban environmental conditions.

As we navigate the fast-paced era of urban expansion, the integration of machine learning (ML) and remote sensing (RS) has become a cornerstone in environmental management. This research, focusing on Silchar City, a non-attainment city under the National Clean Air Program (NCAP), leverages these advanced technologies to understand the urban microclimate and its implications on the health, resilience, and sustainability of the built environment. The rise in land surface temperature (LST) and changes in land use and land cover (LULC) have been identified as key contributors to thermal dynamics, particularly focusing on the development of urban heat islands (UHIs). The Urban Thermal Field Variance Index (UTFVI) can assess the influence of UHIs, which is considered a parameter for ecological quality assessment. This research examines the interlinkages among urban expansion, LST, and thermal dynamics in Silchar City due to a substantial rise in air temperature, poor air quality, and particulate matter PM_2.5. Using Landsat satellite imagery, LULC maps were derived for 2000, 2010, and 2020 by applying a supervised classification approach. LST was calculated by converting thermal band spectral radiance into brightness temperature. We utilized Cellular Automata (CA) and Artificial Neural Networks (ANNs) to project potential scenarios up to the year 2040. Over the two-decade period from 2000 to 2020, we observed a 21% expansion in built-up areas, primarily at the expense of vegetation and agricultural lands. This land transformation contributed to increased LST, with over 10% of the area exceeding 25 °C in 2020 compared with just 1% in 2000. The CA model predicts built-up areas will grow by an additional 26% by 2040, causing LST to rise by 4 °C. The UTFVI analysis reveals declining thermal comfort, with the worst affected zone projected to expand by 7 km². The increase in PM_2.5 and aerosol optical depth over the past two decades further indicates deteriorating air quality. This study underscores the potential of ML and RS in environmental management, providing valuable insights into urban expansion, thermal dynamics, and air quality that can guide policy formulation for sustainable urban planning.