YOLO deep learning algorithm for object detection in agriculture: a review DOI Creative Commons
S. Kanaga Suba Raja, R. Kumaraperumal,

P Pazhanivelan

et al.

Journal of Agricultural Engineering, Journal Year: 2024, Volume and Issue: 55(4)

Published: Dec. 13, 2024

YOLO represents the one-stage object detection also called regression-based detection. Object in given input is directly classified and located instead of using candidate region. The accuracy from two-stage higher than where speed has become popular because its Detection accuracy, good generalization, open-source, speed. boasts exceptional due to approach regression problems for frame detection, eliminating need a complex pipeline. In agriculture, remote sensing drone technologies classifies detects crops, diseases, pests, used land use mapping, environmental monitoring, urban planning, wildlife. Recent research highlights YOLO's impressive performance various agricultural applications. For instance, YOLOv4 demonstrated high counting locating small objects UAV-captured images bean plants, achieving an AP 84.8% recall 89%. Similarly, YOLOv5 showed significant precision identifying rice leaf with rate 90%. this review, we discuss basic principles behind YOLO, different versions limitations, application agriculture farming.

Language: Английский

Advancements in Remote Sensing Imagery Applications for Precision Management in Olive Growing: A Systematic Review DOI Creative Commons
Pedro Marques, Luís Pádua, Joaquim J. Sousa

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(8), P. 1324 - 1324

Published: April 9, 2024

This systematic review explores the role of remote sensing technology in addressing requirements sustainable olive growing, set against backdrop growing global food demands and contemporary environmental constraints agriculture. The critical analysis presented this document assesses different platforms (satellites, manned aircraft vehicles, unmanned aerial vehicles terrestrial equipment) sensors (RGB, multispectral, thermal, hyperspectral LiDAR), emphasizing their strategic selection based on specific study aims geographical scales. Focusing particularly prominent Mediterranean region, article analyzes diverse applications sensing, including management inventory irrigation; detection/monitoring diseases phenology; estimation crucial parameters regarding biophysical parameters, water stress indicators, crop evapotranspiration yield. Through a perspective insights from studies conducted olive-growing regions, underscores potential benefits shaping improving agricultural practices, mitigating impacts ensuring economic viability trees.

Language: Английский

Citations

10
Drone-Assisted Plant Disease Identification Using Artificial Intelligence: A Critical Review DOI Creative Commons
Hicham Slimani, Jamal El Mhamdi, Abdelilah Jilbab

et al.

International Journal of Computing and Digital Systems, Journal Year: 2023, Volume and Issue: 14(1), P. 10433 - 10446

Published: Oct. 20, 2023

Artificial intelligence has been incorporated into modern agriculture to increase agricultural output and resource efficiency.Utilizing deep learning, particularly convolutional neural networks, for recognizing diagnosing plant diseases is tempting.In parallel, drone integration in precision accelerated, providing new potential crop monitoring, map creation, targeted treatments.This study analyzes over 100 significant research articles published between 2018 2023, examining the interaction drones artificial identifying diseases.We begin by explaining value of sensor technology carefully mapping area.The various CNN architectures drone-based approaches essential precise illness detection diagnosis are then highlighted a thorough review.Our highlights how this combination can transform managed completely.This emphasizes conceptual underpinnings fusion, even if fulfilling promise needs additional investigation.In conclusion, we expect changing paths direct improvements field integrate AI, drones, pathology coherent framework with consequences.

Language: Английский

Citations

12
Detection of Cotton Diseases by YOLOv8 on UAV Images Using the RT-DETR Backbone DOI

Zakaria Kinda,

Sadouanouan Malo, Thierry Roger Bayala

et al.

Lecture notes in networks and systems, Journal Year: 2025, Volume and Issue: unknown, P. 3 - 13

Published: Jan. 1, 2025

Language: Английский

Citations

0
A Deep Learning-Based Model for Efficient Olive Leaf Disease Classification DOI
Hajar Hamdaoui, Yassine Zarrouk,

Nour-Eddine Kouddane

et al.

Lecture notes in electrical engineering, Journal Year: 2025, Volume and Issue: unknown, P. 477 - 487

Published: Jan. 1, 2025

Language: Английский

Citations

0
RETRACTED: Advancing disease identification in fava bean crops: A novel deep learning solution integrating YOLO-NAS for precise rust DOI
Hicham Slimani, Jamal El Mhamdi, Abdelilah Jilbab

et al.

Journal of Intelligent & Fuzzy Systems, Journal Year: 2023, Volume and Issue: 46(2), P. 3475 - 3489

Published: Dec. 19, 2023

This article has been retracted. A retraction notice can be found at https://doi.org/10.3233/JIFS-219433.

Language: Английский

Citations

5
Self-Attention-Mechanism-Improved YoloX-S for Briquette Biofuels Object Detection DOI Open Access
Yaxin Wang, Xin‐Yuan Liu,

Fanzhen Wang

et al.

Sustainability, Journal Year: 2023, Volume and Issue: 15(19), P. 14437 - 14437

Published: Oct. 3, 2023

Fuel types are essential for the control systems of briquette biofuel boilers, as optimal combustion condition varies with fuel type. Moreover, use coal in biomass boilers is illegal China, and detection coals will, time, provide effective information environmental supervision. This study established a identification method based on object images, including straw pellets, blocks, wood coal. The YoloX-S model was used baseline network, proposed this improved performance by adding self-attention mechanism module. showed better accuracy than Yolo-L, YoloX-S, Yolov5, Yolov7, Yolov8 models. experimental results regarding show that can effectively distinguish from overcome false missed detections found recognition pellets original YoloX model. However, interference complex background greatly reduce confidence using

Language: Английский

Citations

2
Verticillium wilt of olive and its control caused by the hemibiotrophic soil-borne fungus Verticillium dahliae DOI Creative Commons

Abdelhak Rhouma,

Lobna Hajji-Hedfi, Mohamed El Amine Kouadri

et al.

Microbial Biosystems, Journal Year: 2023, Volume and Issue: 8(2), P. 25 - 36

Published: Dec. 1, 2023

trees worldwide is estimated to be 10.1 million hectares as of 2023 (FAO, 2023; IOC, 2023).While olive are renowned for their resilience, they not impervious the threats posed by plant diseases.Various pathogens, including bacteria and fungi, can cause diseases that have potential devastate production.These lead reduced yields, poor fruit quality, in severe cases, complete tree loss (Acharya et al., 2020;Montes-Osuna & Mercado-Blanco, 2020).One pose a threat Reviews

Language: Английский

Citations

2
Enhancing unmanned aerial vehicle capabilities: integrating YOLO algorithms for diverse industrial applications DOI Creative Commons

Nikolai Guliutin,

Oleslav Antamoshkin

ITM Web of Conferences, Journal Year: 2024, Volume and Issue: 59, P. 03012 - 03012

Published: Jan. 1, 2024

The integration of UAVs with advanced deep learning algorithms, particularly the You Only Look Once models, has opened new horizons in various industries. This paper explores transformative impact YOLO-based systems across diverse sectors, including agriculture, forest fire detection, ecology, marine science, target and UAV navigation. We delve into specific applications different YOLO ranging from YOLOv3 to lightweight YOLOv8, highlighting their unique contributions enhancing functionalities. In equipped algorithms have revolutionized disease crop monitoring, weed management, contributing sustainable farming practices. application management showcases capability these real-time localization analysis. ecological sciences, use models significantly improved wildlife environmental surveillance, resource management. Target detection studies reveal efficacy processing complex imagery for accurate efficient object recognition. Moreover, advancements navigation, through visual landing recognition operation challenging environments, underscore versatility efficiency integrated systems. comprehensive analysis demonstrates profound technologies fields, underscoring potential future innovations applications.

Language: Английский

Citations

0
Identification of Araucaria angustifolia trees in an urban forest fragment using UAV images and YOLOv7 structure DOI Creative Commons
Alan D’Oliveira Correa, Matheus Kopp Prandini, Vinícius Costa Cysneiros

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: April 12, 2024

Abstract This study addresses the identification of individuals Araucaria angustifolia species in urban forest fragments, specifically Mixed Ombrophilous Forest (FOM) Curitiba, Paraná, Brazil. The aim is to use UAV images and computer vision technique YOLOv7 model detect A. angustifolia. FOM essential for local biodiversity conservation human well-being but faces challenges due sprawl conversion land agriculture. critically endangered, requiring actions strategies its conservation. highlights role Unmanned Aerial Vehicles (UAVs) deep learning techniques, such as Convolutional Neural Networks (CNNs), identifying tree ecosystems. YOLOv7, an architecture based on CNNs, was chosen because detection capacity. especially effective at detecting a wide variety objects, including people, vehicles, animals, household road signs much more, making it ideal choice environments. data obtained by DJI Mavic 3 UAV. Utilizing UAV, area flown over, generating orthomosaic that subsequently divided into 14 parts training, validation, testing. trained with trees present area. results show achieved precision 79.3%, recall 86.8%, Mean Average Precision 87% during training. Comparative analysis inventory reveals promising performance trees. average confidence model's classification 76.18 ± 12.88%, 80.81% being most frequent median result. uses integration technology, assess approach provides important tool aimed assessing managing remnants.

Language: Английский

Citations

0
Development of a Drone-Based Phenotyping System for European Pear Rust (Gymnosporangium sabinae) in Orchards DOI Creative Commons
Virginia Maß,

Johannes Seidl-Schulz,

M. Leipnitz

et al.

Agronomy, Journal Year: 2024, Volume and Issue: 14(11), P. 2643 - 2643

Published: Nov. 9, 2024

Computer vision techniques offer promising tools for disease detection in orchards and can enable effective phenotyping the selection of resistant cultivars breeding programmes research. In this study, a digital system monitoring was developed using drones, object photogrammetry, focusing on European pear rust (Gymnosporangium sabinae) as model pathogen. High-resolution RGB images from ten low-altitude drone flights were collected 2021, 2022 2023. A total 16,251 annotations leaves with symptoms created 584 Vision Annotation Tool (CVAT). The YOLO algorithm used automatic symptoms. novel photogrammetric approach Agisoft’s Metashape Professional software ensured accurate localisation geographic information QGIS calculated infestation intensity per tree based canopy areas. This drone-based shows results could considerably simplify tasks involved fruit

Language: Английский

Citations

0