Deep artificial intelligence applications for natural disaster management systems: A methodological review

Ecological Indicators, Journal Year: 2024, Volume and Issue: 163, P. 112067 - 112067

Published: May 6, 2024

Deep learning techniques through semantic segmentation networks have been widely used for natural disaster analysis and response. The underlying base of these implementations relies on convolutional neural (CNNs) that can accurately precisely identify locate the respective areas interest within satellite imagery or other forms remote sensing data, thereby assisting in evaluation, rescue planning, restoration endeavours. Most CNN-based deep-learning models encounter challenges related to loss spatial information insufficient feature representation. This issue be attributed their suboptimal design layers capture multiscale-context failure include optimal during pooling procedures. In early CNNs, network encodes elementary representations, such as edges corners, whereas, progresses toward later layers, it more intricate characteristics, complicated geometric shapes. theory, is advantageous a extract features from several levels because generally yield improved results when both simple maps are employed together. study comprehensively reviews current developments deep methodologies segment images associated with disasters. Several popular models, SegNet U-Net, FCNs, FCDenseNet, PSPNet, HRNet, DeepLab, exhibited notable achievements various applications, including forest fire delineation, flood mapping, earthquake damage assessment. These demonstrate high level efficacy distinguishing between different land cover types, detecting infrastructure has compromised damaged, identifying regions fire-susceptible further dangers.

Language: Английский

---

Integrating Artificial Intelligence Agents with the Internet of Things for Enhanced Environmental Monitoring: Applications in Water Quality and Climate Data

Electronics, Journal Year: 2025, Volume and Issue: 14(4), P. 696 - 696

Published: Feb. 11, 2025

The integration of artificial intelligence (AI) agents with the Internet Things (IoT) has marked a transformative shift in environmental monitoring and management, enabling advanced data gathering, in-depth analysis, more effective decision making. This comprehensive literature review explores AI IoT technologies within sciences, particular focus on applications related to water quality climate data. methodology involves systematic search selection relevant studies, followed by thematic, meta-, comparative analyses synthesize current research trends, benefits, challenges, gaps. highlights how enhances IoT’s collection capabilities through predictive modeling, real-time analytics, automated making, thereby improving accuracy, timeliness, efficiency systems. Key benefits identified include enhanced precision, cost efficiency, scalability, facilitation proactive management. Nevertheless, this encounters substantial obstacles, including issues quality, interoperability, security, technical constraints, ethical concerns. Future developments point toward enhancements technologies, incorporation innovations like blockchain edge computing, potential formation global systems, greater public involvement citizen science initiatives. Overcoming these challenges embracing new technological trends could enable play pivotal role strengthening sustainability resilience.

Language: Английский

---

Quantitative regularization in robust vision transformer for remote sensing image classification

The Photogrammetric Record, Journal Year: 2024, Volume and Issue: 39(186), P. 340 - 372

Published: April 24, 2024

Abstract Vision Transformers (ViTs) are exceptional at vision tasks. However, when applied to remote sensing images (RSIs), existing methods often necessitate extensive modifications of ViTs rival convolutional neural networks (CNNs). This requirement significantly impedes the application in geosciences, particularly for researchers who lack time comprehensive model redesign. To address this issue, we introduce concept quantitative regularization (QR), designed enhance performance RSI classification. QR represents an effective algorithm that adeptly manages domain discrepancies RSIs and can be integrated with any transfer learning. We evaluated effectiveness using three ViT architectures: vanilla ViT, Swin‐ViT Next‐ViT, on four datasets: AID30, NWPU45, AFGR50 UCM21. The results reveal our Next‐ViT surpasses 39 other advanced published past 3 years, maintaining robust even a limited number training samples. also discovered achieve higher accuracy robustness compared same backbone. Our findings confirm as CNNs classification, regardless dataset size. approach exclusively employs open‐source easily accessible strategies. Consequently, believe method lower barriers geoscience intending use applications.

Language: Английский

---

Towards the next generation of Geospatial Artificial Intelligence

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2025, Volume and Issue: 136, P. 104368 - 104368

Published: Jan. 20, 2025

Language: Английский

---

Remote sensing of cyanobacterial harmful algal blooms: Current trends and future directions

Progress in Environmental Geography, Journal Year: 2025, Volume and Issue: 4(1), P. 131 - 150

Published: March 1, 2025

Cyanobacterial harmful algal blooms (CyanoHABs) pose significant threats to aquatic ecosystems, public health, and economic sustainability worldwide. This progress report explores recent advancements in CyanoHAB detection, quantification, monitoring using multi-sensor remote sensing approaches, artificial intelligence (AI) applications, their integration with health impact studies. We presented the capabilities of various satellite sensors CyanoHABs across different spatial temporal scales, discussing multiple data sources overcome individual sensor limitations. The highlights promise AI, particularly machine learning (ML) techniques, improving detection forecasting, demonstrating how ML methods consistently outperformed traditional algorithms estimating phycocyanin concentrations, a key indicator CyanoHABs.We examined development cloud-based applications for real-time awareness. Furthermore, we explored impacts on humans animals, emphasizing role mitigating these effects. implications CyanoHAB-related issues are discussed, along potential integrating epidemiological Overall, this underscores importance cross-disciplinary, integrated approaches that combine cutting-edge technologies, advanced assessments address complex challenges posed by inland waters.

Language: Английский

---

Synergy between Artificial Intelligence and Hyperspectral Imagining—A Review

Technologies, Journal Year: 2024, Volume and Issue: 12(9), P. 163 - 163

Published: Sept. 13, 2024

The synergy between artificial intelligence (AI) and hyperspectral imaging (HSI) holds tremendous potential across a wide array of fields. By leveraging AI, the processing interpretation vast complex data generated by HSI are significantly enhanced, allowing for more accurate, efficient, insightful analysis. This powerful combination has to revolutionize key areas such as agriculture, environmental monitoring, medical diagnostics providing precise, real-time insights that were previously unattainable. In instance, AI-driven can enable precise crop monitoring disease detection, optimizing yields reducing waste. this technology track changes in ecosystems with unprecedented detail, aiding conservation efforts disaster response. diagnostics, AI-HSI could earlier accurate improving patient outcomes. As AI algorithms advance, their integration is expected drive innovations enhance decision-making various sectors. continued development these technologies likely open new frontiers scientific research practical applications, accessible tools wider range users.

Language: Английский

---

Estimation of Maize Residue Cover Using Remote Sensing Based on Adaptive Threshold Segmentation and CatBoost Algorithm

Agriculture, Journal Year: 2024, Volume and Issue: 14(5), P. 711 - 711

Published: April 30, 2024

Maize residue cover (MRC) is an important parameter to quantify the degree of crop in field and its spatial distribution characteristics. It also a key indicator conservation tillage. Rapid accurate estimation maize mapping are great significance increasing soil organic carbon, reducing wind water erosion, maintaining water. Currently, large areas suffers from low modeling accuracy poor working efficiency. Therefore, how improve efficiency has become research hotspot. In this study, adaptive threshold segmentation (Yen) CatBoost algorithm integrated fused construct coverage method based on multispectral remote sensing images. The planting around Sihe Town Jilin Province, China, were selected as typical experimental regions, unmanned aerial vehicle (UAV) was employed capture images sample plots within area. Yen applied calculate analyze cover. successive projections (SPA) used extract spectral feature indices Sentinel-2A Subsequently, model indices, thereby plotting map results show that image outperforms traditional methods, with highest Dice coefficient reaching 81.71%, effectively improving recognition plots. By combining index calculation SPA algorithm, features extracted, such NDTI STI determined. These significantly correlated built using surpasses machine learning models, maximum determination (R2) 0.83 validation set. constructed algorithms enhances reliability estimating imagery, providing reliable data support services for precision agriculture

Language: Английский

---

Artificial intelligence for predicting urban heat island effect and optimising land use/land cover for mitigation: Prospects and recent advancements

Urban Climate, Journal Year: 2024, Volume and Issue: 55, P. 101976 - 101976

Published: May 1, 2024

Rocketing global urbanisation has caused an increase in the Urban Heat Island (UHI) effect, resulting various negative implications for urban environment. Quantifying Surface UHI (SUHI) effect using Land Temperature (LST), Local Climate Zones (LCZ), and deep learning algorithms such as Convolutional Neural Networks (CNN) pix2pix have prospects aiding sustainable city planning modification. Most research on mitigating SUHI promotes greenery a solution, allowing LCZ optimisation to be explored. Using Vulnerability Index (HVI) evolutionary like Genetic Algorithms (GA) Particle Swarm Optimisation (PSO) show promise achieving high-quality solutions. This short communication explores potential of these artificial intelligence technologies combat enhance sustainability.

Language: Английский

---

Mitigating Climate Change

Practice, progress, and proficiency in sustainability, Journal Year: 2024, Volume and Issue: unknown, P. 161 - 200

Published: Aug. 27, 2024

The existential threat presented by climate change demands an unprecedented response. Existing environmental regulations are insufficient for the pollution concerns that arise from our complicated and integrated global economy. AI has potential to completely revolutionize existing regulatory frameworks dramatically improve mitigation with superior data collection, modeling & new enforcement capabilities. Using a doctrinal approach, it studied both national international laws found best practices as well legal obstacles, such need privacy algorithmic bias concerns. It discovered health law regulation compliance of in public health. concluded artificial intelligence had vastly partially but theoretically, strict can curb worst impulses unscrupulous AI. recommended policymakers collaborate experts researchers ensure quality action.

Language: Английский

---

Marine Equipment Siting Using Machine-Learning-Based Ocean Remote Sensing Data: Current Status and Future Prospects

Sustainability, Journal Year: 2024, Volume and Issue: 16(20), P. 8889 - 8889

Published: Oct. 14, 2024

As the global climate changes, there is an increasing focus on oceans and their protection exploitation. However, exploration of necessitates construction marine equipment, siting such equipment has become a significant challenge. With ongoing development computers, machine learning using remote sensing data proven to be effective solution this problem. This paper reviews history technology, introduces conditions required for site selection through measurement analysis, uses cluster analysis methods identify areas as research hotspot ocean sensing. The aims integrate into Through review discussion article, limitations shortcomings current stage are identified, relevant proposals put forward.

Language: Английский

---