Integration of Remote Sensing and Machine Learning for Precision Agriculture: A Comprehensive Perspective on Applications

Agronomy, Journal Year: 2024, Volume and Issue: 14(9), P. 1975 - 1975

Published: Sept. 1, 2024

Due to current global population growth, resource shortages, and climate change, traditional agricultural models face major challenges. Precision agriculture (PA), as a way realize the accurate management decision support of production processes using modern information technology, is becoming an effective method solving these In particular, combination remote sensing technology machine learning algorithms brings new possibilities for PA. However, there are relatively few comprehensive systematic reviews on integrated application two technologies. For this reason, study conducts literature search Web Science, Scopus, Google Scholar, PubMed databases analyzes in PA over last 10 years. The found that: (1) because their varied characteristics, different types data exhibit significant differences meeting needs PA, which hyperspectral most widely used method, accounting more than 30% results. UAV offers greatest potential, about 24% data, showing upward trend. (2) Machine displays obvious advantages promoting development vector algorithm 20%, followed by random forest algorithm, 18% methods used. addition, also discusses main challenges faced currently, such difficult problems regarding acquisition processing high-quality model interpretation, generalization ability, considers future trends, intelligence automation, strengthening international cooperation sharing, sustainable transformation achievements. summary, can provide ideas references combined with promote

Language: Английский

---

Editorial: Transformer Models for Multi-source Visual Fusion and Understanding

Information Fusion, Journal Year: 2025, Volume and Issue: unknown, P. 103112 - 103112

Published: March 1, 2025

Language: Английский

---

Research on Super-Resolution Reconstruction Algorithms for Remote Sensing Images of Coastal Zone Based on Deep Learning

Land, Journal Year: 2025, Volume and Issue: 14(4), P. 733 - 733

Published: March 29, 2025

High-resolution multispectral remote sensing imagery is widely used in critical fields such as coastal zone management and marine engineering. However, obtaining images at a low cost remains significant challenge. To address this issue, we propose the MRSRGAN method (multi-scale residual super-resolution generative adversarial network). The leverages Sentinel-2 GF-2 imagery, selecting nine typical land cover types zones, constructs small sample dataset containing 5210 images. extracts differential features between high-resolution (HR) low-resolution (LR) to generate In our approach, design three key modules: fusion attention-enhanced module (FAERM), multi-scale attention (MSAF), feature extraction (MSFE). These modules mitigate gradient vanishing extract image different scales enhance reconstruction. We conducted experiments verify their effectiveness. results demonstrate that approach reduces Learned Perceptual Image Patch Similarity (LPIPS) by 14.34% improves Structural Index (SSIM) 11.85%. It effectively issue where large-scale span of ground objects makes single-scale convolution insufficient for capturing detailed features, thereby improving restoration effect details significantly enhancing sharpness object edges.

Language: Английский

---

A lightweight multi-scale feature fusion method for detecting defects in water-based wood paint surfaces

Measurement, Journal Year: 2025, Volume and Issue: unknown, P. 117505 - 117505

Published: April 1, 2025

Language: Английский

---

Detection of the stem-boring damage by pine shoot beetle (Tomicus spp.) to Yunan pine (Pinus yunnanensis Franch.) using UAV hyperspectral data

Frontiers in Plant Science, Journal Year: 2025, Volume and Issue: 16

Published: April 9, 2025

The stem-boring damage caused by pine shoot beetle (PSB, Tomicus spp.) cuts off the transmission of water and nutrients. aggregation beetles during stage results in rapid mortality Yunnan pines (Pinus yunnanensis Franch.). Timely identification precise localization PSB are crucial for removing infected wood preventing further spread infestation. Unmanned airborne vehicle (UAV) hyperspectral data demonstrate great potential assessing pest outbreaks forested landscapes. However, there is a lack studies investigating application accuracy UAV detecting damage. In this study, we compared differences spectral features healthy (H level), three levels shoot-feeding (E, M S levels), (T then used Random Forest (RF) algorithm PSB. specific canopy features, including red edge (such as Dr, SDr, D711), blue Db SDb), chlorophyll-related indices (e.g., MCARI) were sensitive to RF models showed that first-order derivative (FD) (SIs) played an important role detection. Models incorporating FD bands, SIs combination all variables proved more effective These findings damage, which significantly contributed prevention management infestations.

Language: Английский

---

WMC-RTDETR: a lightweight tea disease detection model

Frontiers in Plant Science, Journal Year: 2025, Volume and Issue: 16

Published: May 6, 2025

Tea pest and disease detection is crucial in tea plantation management, however, challenges such as multi-target occlusion complex background impact accuracy efficiency. To address these issues, this paper proposes an improved lightweight model, WMC-RTDETR, based on the RT-DETR model. The model significantly enhances ability to capture multi-scale features by introducing wavelet transform convolution, improving feature extraction backgrounds, increasing efficiency while reducing number of parameters. Combined with multiscale multihead self-attention, global fusion across scales realized, which effectively overcomes shortcomings traditional attention mechanisms small target detection. Additionally, a context-guided spatial reconstruction pyramid network designed refine through contextual information, thereby robustness scenes. Experimental results show that proposed achieves 97.7% 83.1% respectively mAP50 mAP50:95 indicators, outperform original In addition, parameters floating-point operations are reduced 35.48% 40.42% respectively, enabling highly efficient accurate pests diseases scenarios. Furthermore, successfully deploys Raspberry Pi platform, proves it has good real-time performance resource-constrained embedded environments, providing practical solution for low-cost monitoring agricultural

Language: Английский

---

An improved SSH optimization model for wireless security in Complex networks

2022 13th International Conference on Computing Communication and Networking Technologies (ICCCNT), Journal Year: 2024, Volume and Issue: unknown, P. 1 - 7

Published: June 24, 2024

Language: Английский

---

YOLOv8-RD: High-Robust Pine Wilt Disease Detection Method Based on Residual Fuzzy YOLOv8

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Journal Year: 2024, Volume and Issue: 18, P. 385 - 397

Published: Nov. 11, 2024

Language: Английский

---

A Lightweight Network for Concrete Structure Crack Detection in Complex Environments Based on Cloud-Edge Collaboration

Published: Jan. 1, 2024

Language: Английский

---

OmniFuse: A general modality fusion framework for multi-modality learning on low-quality medical data

Information Fusion, Journal Year: 2024, Volume and Issue: unknown, P. 102890 - 102890

Published: Dec. 1, 2024

Language: Английский

---