Integrating IoT and Image Processing for Crop Monitoring: A LoRa-Based Solution for Citrus Pest Detection DOI Open Access

Joel L. Quispe-Vilca,

Edison Moreno-Cardenas, Erwin J. Sacoto-Cabrera

et al.

Electronics, Journal Year: 2024, Volume and Issue: 13(24), P. 4863 - 4863

Published: Dec. 10, 2024

Today, agriculture faces many challenges, such as the use of inefficient methods that affect crop quality. Precision (PA), combined with advanced technologies, improves monitoring, while integration wireless communication optimizes processes and resources. This work presents design a prototype applied in precision agriculture, which allows acquisition, processing, transmission information extracted from Cotonet pest to The Things Network (TTN) cloud server. integrates technologies protocols LoRaWAN, Message Queuing Telemetry Transport (MQTT), Internet (IoT) sensors, Computer Vision. employs robust processing segmentation algorithm, recognition pests citrus plants based on color. results show lighting conditions, weather, time day influence quality captured images. relationship between image resolution, brightness, shows higher-resolution images (1920 × 1080 pixels per image) provide better detection (greater than 50% index) but require longer (28.415 ms average). Furthermore, developed system effectively detects an index affection Planococcus citri (Cotonet) agricultural plantations through end-to-end technological implementation communication, IoT technologies.

Language: Английский

An intelligent multi-modal neural framework for accurate fruit grading localization and yield estimation DOI
Ghassan Faisal Albaaji, S. S. Vinod Chandra, Misaj Sharafudeen

et al.

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 126366 - 126366

Published: Jan. 1, 2025

Language: Английский

Citations

1
The Lightweight Deep Learning Model in Sunflower Disease Identification: A Comparative Study DOI Creative Commons
Liqian Zhang, Xiao Wu

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(4), P. 2104 - 2104

Published: Feb. 17, 2025

(1) With the development of artificial intelligence, people expect to use modern information technology solve critical problems encountered in agriculture. How identify sunflower diseases as early and quickly possible take corresponding measures has become a key issue for increasing crop production farmers’ income. Sunflowers, an important oil crop, are vulnerable infections by various diseases, such downy mildew, leaf scar, gray mold, etc. (2) In order select better lightweight model that can be embedded into mobile devices or disease detection, we compared five deep learning models this study, including SqueezeNet, ShuffleNetV2, MnasNet-A1, MobileNetV3-Small, EfficientNetV2-Small. The dataset used train test included 1892 images. These images were divided four categories, namely, fresh leaves. (3) By evaluating accuracy, precision, recall, F1 score each model, found EfficeintNetV2-Small exhibited highest performance with accuracy 90.19%. Whereas other models, achieved accuracies 84.08%, 79.31%, 88.59%, respectively. To address problem poor generalization ability caused small datasets, adopted transfer technique. After doing that, recognition EfficeintNetV2-Small, reached 96.02%, 95.23%, 94.96%, 96.92%, 99.20%, these improved 14.2%, 20%, 7.2%, 15.2%, 10%. Based on comparative results, was optimal choice identification due its high detection accuracy.

Language: Английский

Citations

1
Digital transformation of the agri-food system DOI Creative Commons
Mahdi Vahdanjoo, Claus Aage Grøn Sørensen, Michael Nørremark

et al.

Current Opinion in Food Science, Journal Year: 2025, Volume and Issue: unknown, P. 101287 - 101287

Published: Feb. 1, 2025

Language: Английский

Citations

1
DeepLeaf: an optimized deep learning approach for automated recognition of grapevine leaf diseases DOI Creative Commons
Fatma M. Talaat, Mahmoud Y. Shams, Samah A. Gamel

et al.

Neural Computing and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 28, 2025

Language: Английский

Citations

1
SMART DAIRY FARMING: ENHANCED EFFICIENCY, PRODUCTIVITY AND ANIMAL WELFARE THROUGH THE INTERNET OF THINGS AND CLOUD INTEGRATION DOI Creative Commons
Manzar Abbas, Ghulam Abbas,

S. Jaffery

et al.

The Journal of Animal and Plant Sciences, Journal Year: 2025, Volume and Issue: 1, P. 18 - 35

Published: Jan. 8, 2025

Dairy industry faces numerous challenges today and, in the future, including labor shortage, stemming from economic pressure due to high cost and insufficient returns, evolving marketing dynamics. In order cope with these challenges, integration of advance technologies such as automation data analytics is indispensable. The Internet Things (IoT) has enabled development “smart” devices installed sensors smart collars, wearables, thermometer, hygrometer, air quality detectors for efficient sustainable dairy farming. Moreover, vast volume generated by IoT necessitates cloud computing effective handling. However, this presents challenges; particular, overload superfluous communication noise. To address this, pre-processing trimming services gateways, networks, fog have been employed. livestock farming, CoT revolutionized real-time monitoring, advanced care, in-time ovum pick-up, vitro fertilization, embryo transfer, artificial insemination, milk production, gene selection. Through sensors, regarding an animal’s health (e.g., body temperature, level reproductive hormones, vaginal pH), behavior, environment facilitated animal welfare practices. CoT’s cloud-based infrastructure enables comprehensive analysis, leading improved veterinary early disease detection, insightful research into diverse species’ Ultimately, signify a paradigm shift transcending mere offer holistic, data-driven approach that harmonizes productivity welfare. By leveraging innovations, sector poised achieve growth saving 178% on feed pushing, 44.05% milking, 121.97% cleansing, 126.2% herd 109.3% analyzing forecasting. This study falls under umbrella UNO’s goals development. Keywords: Things, computing, intelligent breeding, management, farm management

Language: Английский

Citations

0
Plant Leaf Disease Detection Using Deep Learning: A Multi-Dataset Approach DOI Creative Commons

Manjunatha Shettigere Krishna,

Pedro Machado, Richard I. Otuka

et al.

J — Multidisciplinary Scientific Journal, Journal Year: 2025, Volume and Issue: 8(1), P. 4 - 4

Published: Jan. 15, 2025

Agricultural productivity is increasingly threatened by plant diseases, which can spread rapidly and lead to significant crop losses if not identified early. Detecting diseases accurately in diverse uncontrolled environments remains challenging, as most current detection methods rely heavily on lab-captured images that may generalise well real-world settings. This paper aims develop models capable of identifying across conditions, overcoming the limitations existing methods. A combined dataset was utilised, incorporating PlantDoc with web-sourced plants from online platforms. State-of-the-art convolutional neural network (CNN) architectures, including EfficientNet-B0, EfficientNet-B3, ResNet50, DenseNet201, were employed fine-tuned for leaf disease classification. key contribution this work application enhanced data augmentation techniques, such adding Gaussian noise, improve model generalisation. The results demonstrated varied performance datasets. When trained tested dataset, EfficientNet-B3 achieved an accuracy 73.31%. In cross-dataset evaluation, where a reached 76.77% accuracy. best combination PlanDoc datasets resulting 80.19% indicating very good generalisation conditions. Class-wise F1-scores consistently exceeded 90% apple rust grape all models, demonstrating effectiveness approach detection.

Language: Английский

Citations

0
Research on the Development and Application of an Intelligent Aquaculture System DOI Creative Commons

Yongquan Nie,

Huaining Yang,

Keming Qu

et al.

Information Resources Management Journal, Journal Year: 2025, Volume and Issue: 38(1), P. 1 - 20

Published: Feb. 12, 2025

This research addresses the pressing need for sustainable practices in aquaculture, which faces challenges, like environmental degradation. The study aims to evaluate effectiveness of an intelligent aquaculture system (IAS) improving key performance indicators shrimp farming. Methodologically, it focuses on a specific farm divided into 10 breeding zones, with number 3 area selected experimentation. Data parameters and metrics were collected comparative analysis against traditional practices. Results showed significant improvements: IAS achieved feed conversion rate 90.22% growth 50 g/week, outperforming methods. Additionally, exhibited lower disease incidence mortality rates, indicating enhanced safety. concludes that IASs can substantially improve operational efficiency sustainability, offering valuable insights future

Language: Английский

Citations

0
Sustainable Agriculture: Integrating IoT and AI for Resource-Efficient Farming DOI
Franciskus Antonius Alijoyo,

Shamim Ahmad Khan,

Deepak Gupta

et al.

Published: Jan. 7, 2025

Language: Английский

Citations

0
Multi-Flight Path Planning for a Single Agricultural Drone in a Regular Farmland Area DOI Open Access
Huijuan Dong, Xiaohan Ma, Si Zhang

et al.

Sustainability, Journal Year: 2025, Volume and Issue: 17(6), P. 2433 - 2433

Published: March 10, 2025

The sustainable management of agricultural systems is crucial for ensuring food security and environmental stewardship. This paper advances development in the field agriculture by focusing on application plant protection drone technology efficiently controlling crop diseases pests. investigates multi-flight path planning a single regular farmland, establishing model that takes into account factors movement characteristics drone. By conducting quantitative analysis farmland information, this optimizes traversal drones two dimensions: pesticide consumption energy consumption. introduces novel optimization algorithm grid activity values adjusting function, based comprehensive coverage planning, dynamically adjusts cost function A* with varying weights. experimental results indicate improved has achieved significant enhancements terms return length efficiency compared to traditional methods. study proposes an efficient method drones, which aids reducing enhancing production efficiency, thereby promoting production.

Language: Английский

Citations

0
Digital Literacy for Sustainability DOI
Mustafa Kayyali

Advances in computational intelligence and robotics book series, Journal Year: 2025, Volume and Issue: unknown, P. 197 - 224

Published: March 14, 2025

As the world faces more complex sustainability concerns, role of digital literacy in influencing students' abilities to contribute a sustainable future has never been vital. The rapid breakthroughs Artificial Intelligence (AI) have created opportunities and difficulties education, requiring integration with education. This article investigates convergence sustainability, claiming that preparing students for an AI-driven society is vital equipping them skills necessary address global concerns. It studies how may be harnessed achieve development, including ability use tools, analyze data, engage critically technology. study also explores pedagogical techniques can implemented integrate into curriculum, providing information competencies they need navigate AI-powered future.

Language: Английский

Citations

0