Internet of Things and Wireless Sensor Networks for Smart Agriculture Applications: A Survey

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 145813 - 145852

Published: Jan. 1, 2023

The increasing food scarcity necessitates sustainable agriculture achieved through automation to meet the growing demand. Integrating Internet of Things (IoT) and Wireless Sensor Networks (WSNs) is crucial in enhancing production across various agricultural domains, encompassing irrigation, soil moisture monitoring, fertilizer optimization control, early-stage pest crop disease management, energy conservation. application protocols such as ZigBee, WiFi, SigFox, LoRaWAN are commonly employed collect real-time data for monitoring purposes. Embracing advanced technology imperative ensure efficient annual production. Therefore, this study emphasizes a comprehensive, future-oriented approach, delving into IoT-WSNs, wireless network protocols, their applications since 2019. It thoroughly discusses overview IoT WSNs, architectures summarization protocols. Furthermore, addresses recent issues challenges related IoT-WSNs proposes mitigation strategies. provides clear recommendations future, emphasizing integration aiming contribute future development smart systems.

Language: Английский

---

PAMICRM: Improving Precision Agriculture Through Multimodal Image Analysis for Crop Water Requirement Estimation Using Multidomain Remote Sensing Data Samples

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 52815 - 52836

Published: Jan. 1, 2024

The growing necessity for sustainable agriculture in the face of escalating global food demands and climate change underscores importance optimizing crop water usage. Current methodologies estimating requirements, primarily relying on traditional remote sensing empirical models, limitations precision, adaptability to diverse types, dynamic environmental conditions. This paper introduces an innovative machine learning model designed accurately estimate requirements optimize irrigation scheduling using data samples. model's robustness stems from its sophisticated pre-processing technique, employing a Grey Wolf Optimized (GWO)-based Non-Local Means (NLM) algorithm denoising images sourced various technologies. Subsequently, these refined undergo transformation into multidomain features through integration Vision Transformer, Fourier, Entropy, Color Map, Gabor Maps. multi-faceted approach ensures comprehensive analysis data, capturing intricate details pertinent requirements. A key innovation our is implementation Dual Elephant Herding Optimization (DEHO) Process. optimizer adeptly selects by minimizing intra-class differences maximizing inter-class variance, thereby enhancing discriminative capabilities. selected are then classified distinct schedule classes Deep Dyna Q Graph Convolutional Network (DDQGCN). network not only augments efficiency classifying sources but also bolsters capability adapt varying types efficacy proposed further augmented incorporating vector autoregressive moving average with eXogenous inputs (VARMAx) algorithms. addition enables conversion output temporal features, facilitating precise predictions soil quality adjustment schedules. Empirical testing this across different geographies mango, rice, cotton crops demonstrated significant improvements over existing methods. exhibited 8.5% increase 8.3% accuracy, 4.9% recall, 5.5% Area Under Curve (AUC), 4.5% specificity, alongside 3.5% reduction response delays. Particularly noteworthy enhanced pre-emption capabilities, reflected 2.9% higher precision pre-emption, 2.5% greater 1.5% increased 2.4% AUC improved 1.9% delay compared methodologies.

Language: Английский

---

A Lightweight and Dynamic Feature Aggregation Method for Cotton Field Weed Detection Based on Enhanced YOLOv8

Electronics, Journal Year: 2024, Volume and Issue: 13(11), P. 2105 - 2105

Published: May 29, 2024

Weed detection is closely related to agricultural production, but often faces the problems of leaf shading and limited computational resources. Therefore, this study proposes an improved weed algorithm based on YOLOv8. Firstly, Dilated Feature Integration Block designed improve feature extraction in backbone network by introducing large kernel convolution multi-scale dilation convolution, which utilizes information from different scales levels. Secondly, solve problem a number parameters fusion process Path Aggregation Pyramid Network, new architecture interaction designed, achieves high-level semantic guide low-level through attention mechanism. Finally, we propose Dynamic Head that YOLOv8 head cannot dynamically focus important features. Comprehensive experiments two publicly accessible datasets show proposed model outperforms benchmark model, with mAP50 mAP75 improving 4.7% 5.0%, 5.3% 3.3%, respectively, whereas only 6.62 M. This illustrates utility potential for cotton fields, marking significant advancement artificial intelligence agriculture.

Language: Английский

---

Improved Weed Detection in Cotton Fields Using Enhanced YOLOv8s with Modified Feature Extraction Modules

Symmetry, Journal Year: 2024, Volume and Issue: 16(4), P. 450 - 450

Published: April 7, 2024

Weed detection plays a crucial role in enhancing cotton agricultural productivity. However, the process is subject to challenges such as target scale diversity and loss of leaf symmetry due shading. Hence, this research presents an enhanced model, EY8-MFEM, for detecting weeds fields. Firstly, ALGA module proposed, which combines local global information feature maps through weighting operations better focus on spatial maps. Following this, C2F-ALGA was developed augment extraction capability underlying backbone network. Secondly, MDPM proposed generate attention matrices by capturing horizontal vertical maps, reducing duplicate Finally, we will replace upsampling YOLOv8 with CARAFE provide performance. Extensive experiments two publicly available datasets showed that F1, mAP50 mAP75 metrics improved 1.2%, 5.1%, 2.9% 3.8%, 1.3%, 2.2%, respectively, compared baseline model. This study showcases algorithm’s potential practical applications weed within fields, promoting significant development artificial intelligence field agriculture.

Language: Английский

---

Issues of complex research in digitalization of adaptive viticulture when implementing artificial intelligence tools

Horticulture and viticulture, Journal Year: 2025, Volume and Issue: 2, P. 39 - 47

Published: May 10, 2025

In recent years, digital technologies and modern computer capable of processing large amounts data have been intensively adopted. At present, specialists deal with the latest technological advances, including neural networks, Internet Things, artificial intelligence. This creates need to consider prospects introducing intelligence in scientific process production viticulture industry, as well highlight problems associated its use. paper reviews elements agricultural production. Over last 15 this direction has received insufficient attention literature, such publications field not exceeding 2 %. requires intensification research on adoption agriculture, viticulture, particularly Russian Federation. Lagging behind issue can lead dependence foreign developers products, which threatens loss independent development unique algorithms control processes plant development, influence abiotic biotic factors, selection optimal ensuring sustainable agriculture general particular. addition, use tools may pose bioterrorism threats through covertly decreasing productivity soil fertility by promotion protection products growth regulators, ignore domestic import-substituting production, equipment, fertilizers. regard, implement cooperative programs accumulate material develop software eliminate above issues is substantiated.

Language: Английский

---

PROBLEMS AND PROSPECTS OF DIGITALIZATION OF RUSSIA'S AGRICULTURAL SECTOR UNDER INTERNATIONAL SANCTIONS

Moscow Economic Journal, Journal Year: 2024, Volume and Issue: unknown, P. 675 - 693

Published: May 24, 2024

The article presents the results of research into problems and prospects digitalization Russia's agro-industrial complex in context international sanctions. Within framework theoretical section, it was established that enterprises sphere is implemented on basis one three main models: pure producer; "from field to counter; ecosystem fork". Russian model AIC has a number specific features: full participation state processes, which both allows retain control over industry simultaneously narrows possibilities free choice digital transformation trajectories for business itself; strategic position high-tech corporations, de facto have competitive advantage tenders and, accordingly, draw large share financing industry; agricultural sector, based state's industry. In author's opinion, complex, taking account impact technological sanctions, are: encouraging holdings develop corporate programs financial support agrotech start-ups; formation national database projects processes categories: precision farming, smart farm, farm management systems, automation sowing disease control; introduction system tax incentives other preferences (e.g., availability preferential loans complexes); breaks use tax-free complexes). revision curricula agrarian universities country with addition disciplines orienting future specialists work technologies.

Language: Английский

---

Farmer’s Touch: Harnessing Technologies to Enhance Crop Production

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

Published: April 8, 2024

Abstract The objective of every agricultural endeavour is to achieve optimal output and productivity under varying conditions, thereby conserving precious resources, energy, reducing production costs. Modern practices focus on the detailed monitoring crop conditions by assessing variables like soil quality, plant vitality, impact fertilizers pesticides, irrigation levels, overall yield. Precision Agriculture defined as a farm management approach that utilizes information technology pinpoint, analyse, address variability found within fields enhance yield, profitability, sustainability, environmental protection, all while cutting down This field employs advanced sensor technologies analytical tools boost support decisions. Our project aims provide guidance both commercial farmers individuals interested in gardening necessary steps precautions for improving comprehensive will offer advice across various sectors, including management, pest control, equipment, marketing, using stack comprises Full-Stack web application, Machine Learning-powered Chatbot integrated via Flask API with translation feature, SQL. primary this deliver virtual blog content web-based assistance enhanced growth

Language: Английский

---

3D Reconstruction of a Tomato Seed Germination Environment Using LiDAR and Color Camera

Published: Oct. 15, 2024

Language: Английский

---

Real-time Environmental Monitoring in Smart Agriculture using TinyML and Machine Learning

Published: Oct. 25, 2024

Language: Английский

---

Artificial Intelligence for All: Challenges and Harnessing Opportunities in AI Democratization

Published: Dec. 12, 2024

Language: Английский

---