Detection of maize tassels for UAV remote sensing image with an improved YOLOX Model DOI Creative Commons

Chaoyu Song,

Fan Zhang,

Jian-sheng LI

et al.

Journal of Integrative Agriculture, Journal Year: 2022, Volume and Issue: 22(6), P. 1671 - 1683

Published: Sept. 24, 2022

Maize tassels detection is essential for future agronomic management in maize planting and breeding, with application yield estimation, growth monitoring, intelligent picking, disease detection, etc. Nevertheless, some problems are gradually becoming more prominent it. shown the field widespread occlusions differ size morphological color of different stages. Aiming at these issues, this study proposes SEYOLOX-tiny model that detects accurately robustness. Firstly, data acquisition method better balanced image quality efficiency obtained images from periods to enrich our dataset by unmanned aerial vehicle (UAV). Moreover, robust network extends YOLOX embedding an attention mechanism realize extraction critical features suppressing noise caused adverse factors (occlusions, overlaps, etc.), which could be suitable operating a complex natural environment. Experimental results verify current work hypothesis show mean average precision ([email protected]) was 95.0%. The [email protected], [email protected], [email protected](area=small), [email protected](area=medium) increased 1.5, 1.8, 5.3, 1.7%, respectively than original model, proposed can meet robustness vision system detection.

Language: Английский

Small Object Detection Algorithm Based on Improved YOLOv8 for Remote Sensing DOI Creative Commons
Yi Hao, Bo Liu, Bin Zhao

et al.

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Journal Year: 2023, Volume and Issue: 17, P. 1734 - 1747

Published: Dec. 5, 2023

Due to the limitations of small targets in remote sensing images such as background noise, poor information, and so on, results commonly used detection algorithms target is not satisfactory. To improve accuracy results, we develop an improved algorithm based on YOLOv8, called LAR-YOLOv8. First, feature extraction network, local module enhanced by using dual-branch architecture attention mechanism, while vision transformer block maximize representation map. Second, attention-guided bi-directional pyramid network designed generate more discriminative information efficiently extracting from shallow through a dynamic sparse adding top-down paths guide subsequent modules for fusion. Finally, RIOU loss function proposed avoid failure shape consistency between predicted ground truth box. Experimental NWPU VHR-10, RSOD CARPK datasets verify that LAR-YOLOv8 achieves satisfactory terms mAP (small), mAP, model parameters FPS, can prove our modifications made original YOLOv8 are effective.

Language: Английский

Citations

79
Recent Advances in Forest Insect Pests and Diseases Monitoring Using UAV-Based Data: A Systematic Review DOI Open Access
André Duarte, Nuno Borralho, Pedro Cabral

et al.

Forests, Journal Year: 2022, Volume and Issue: 13(6), P. 911 - 911

Published: June 10, 2022

Unmanned aerial vehicles (UAVs) are platforms that have been increasingly used over the last decade to collect data for forest insect pest and disease (FIPD) monitoring. These machines provide flexibility, cost efficiency, a high temporal spatial resolution of remotely sensed data. The purpose this review is summarize recent contributions identify knowledge gaps in UAV remote sensing FIPD A systematic was performed using preferred reporting items reviews meta-analysis (PRISMA) protocol. We reviewed full text 49 studies published between 2015 2021. parameters examined were taxonomic characteristics, type sensor, collection pre-processing, processing analytical methods, software used. found number papers on topic has increased years, with most being located China Europe. main FIPDs studied pine wilt (PWD) bark beetles (BB) multirotor architectures. Among sensor types, multispectral red–green–blue (RGB) bands monitoring tasks. Regarding random (RF) deep learning (DL) classifiers frequently applied imagery processing. This paper discusses advantages limitations associated use UAVs methods FIPDs, research challenges presented.

Language: Английский

Citations

75
Ten deep learning techniques to address small data problems with remote sensing DOI Creative Commons
Anastasiia Safonova, Gohar Ghazaryan, Stefan Stiller

et al.

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2023, Volume and Issue: 125, P. 103569 - 103569

Published: Nov. 18, 2023

Researchers and engineers have increasingly used Deep Learning (DL) for a variety of Remote Sensing (RS) tasks. However, data from local observations or via ground truth is often quite limited training DL models, especially when these models represent key socio-environmental problems, such as the monitoring extreme, destructive climate events, biodiversity, sudden changes in ecosystem states. Such cases, also known small pose significant methodological challenges. This review summarises challenges RS domain possibility using emerging techniques to overcome them. We show that problem common challenge across disciplines scales results poor model generalisability transferability. then introduce an overview ten promising techniques: transfer learning, self-supervised semi-supervised few-shot zero-shot active weakly supervised multitask process-aware ensemble learning; we include validation technique spatial k-fold cross validation. Our particular contribution was develop flowchart helps users select which use given by answering few questions. hope our article facilitate applications tackle societally important environmental problems with reference data.

Language: Английский

Citations

68
Review of ground and aerial methods for vegetation cover fraction (fCover) and related quantities estimation: definitions, advances, challenges, and future perspectives DOI
Linyuan Li, Xihan Mu, Hailan Jiang

et al.

ISPRS Journal of Photogrammetry and Remote Sensing, Journal Year: 2023, Volume and Issue: 199, P. 133 - 156

Published: April 12, 2023

Language: Английский

Citations

57
Understanding the cooling capacity and its potential drivers in urban forests at the single tree and cluster scales DOI

Chengcong Wang,

Zhibin Ren, Xinyue Chang

et al.

Sustainable Cities and Society, Journal Year: 2023, Volume and Issue: 93, P. 104531 - 104531

Published: March 16, 2023

Language: Английский

Citations

53
The role of artificial intelligence and digital technologies in dam engineering: Narrative review and outlook DOI
Mohammad Amin Hariri‐Ardebili, Golsa Mahdavi,

Larry K. Nuss

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 126, P. 106813 - 106813

Published: July 25, 2023

Language: Английский

Citations

48
Methods and datasets on semantic segmentation for Unmanned Aerial Vehicle remote sensing images: A review DOI
Jian Cheng, Changjian Deng, Yanzhou Su

et al.

ISPRS Journal of Photogrammetry and Remote Sensing, Journal Year: 2024, Volume and Issue: 211, P. 1 - 34

Published: April 2, 2024

Language: Английский

Citations

24
Comprehensive Review of Drones Collision Avoidance Schemes: Challenges and Open Issues DOI Creative Commons
Mohammad Reza Rezaee, Nor Asilah Wati Abdul Hamid, Masnida Hussin

et al.

IEEE Transactions on Intelligent Transportation Systems, Journal Year: 2024, Volume and Issue: 25(7), P. 6397 - 6426

Published: March 25, 2024

In the contemporary landscape, escalating deployment of drones across diverse industries has ushered in a consequential concern, including ensuring security drone operations. This concern extends to spectrum challenges, encompassing collisions with stationary and mobile obstacles encounters other drones. Moreover, inherent limitations drones, namely constraints on energy consumption, data storage capacity, processing power, present formidable developing collision avoidance algorithms. review paper explores challenges safe operations, focusing avoidance. We explore methods for UAVs from various perspectives, categorizing them into four main groups: obstacle detection avoidance, algorithms, swarm, path optimization. Additionally, our analysis delves machine learning techniques, discusses metrics simulation tools validate systems, delineates local global algorithmic perspectives. Our evaluation reveals significant current prevention Despite advancements, critical UAV network communication are often overlooked, prompting reliance simulation-based research due cost safety concerns. Challenges encompass precise small moving obstacles, minimizing deviations at minimal cost, high automation expenses, prohibitive costs real testbeds, limited environmental comprehension, apprehensions. By addressing these key areas, future can advance field pave way safer more efficient

Language: Английский

Citations

17
Comparative analysis of UAV-based LiDAR and photogrammetric systems for the detection of terrain anomalies in a historical conflict landscape DOI Creative Commons
Marcel Storch, Benjamin Kisliuk, Thomas Jarmer

et al.

Science of Remote Sensing, Journal Year: 2025, Volume and Issue: unknown, P. 100191 - 100191

Published: Jan. 1, 2025

Language: Английский

Citations

2
Harnessing Multi-Source Data and Deep Learning for High-Resolution Land Surface Temperature Gap-Filling Supporting Climate Change Adaptation Activities DOI Creative Commons

Katja Kustura,

Daniel J. Conti,

Matthias Sammer

et al.

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(2), P. 318 - 318

Published: Jan. 17, 2025

Addressing global warming and adapting to the impacts of climate change is a primary focus adaptation strategies at both European national levels. Land surface temperature (LST) widely used proxy for investigating climate-change-induced phenomena, providing insights into radiative properties different land cover types impact urbanization on local characteristics. Accurate continuous estimation across large spatial regions crucial implementation LST as an essential parameter in mitigation strategies. Here, we propose deep-learning-based methodology using multi-source data including Sentinel-2 imagery, cover, meteorological data. Our approach addresses common challenges satellite-derived data, such gaps caused by cloud image border limitations, grid-pattern sensor artifacts, temporal discontinuities due infrequent overpasses. We develop regression-based convolutional neural network model, trained ECOSTRESS (ECOsystem Spaceborne Thermal Radiometer Experiment Space Station) mission which performs pixelwise predictions 5 × patches, capturing contextual information around each pixel. This method not only preserves ECOSTRESS’s native resolution but also fills enhances coverage. In non-gap areas validated against ground truth model achieves with least 80% all pixel errors falling within ±3 °C range. Unlike traditional satellite-based techniques, our leverages high-temporal-resolution capture diurnal variations, allowing more robust time periods. The model’s performance demonstrates potential integrating urban planning, resilience strategies, near-real-time heat stress monitoring, valuable resource assess visualize development use changes.

Language: Английский

Citations

2