Potato Leaf Disease Detection Based on a Lightweight Deep Learning Model DOI Creative Commons

Chao-Yun Chang,

Chih-Chin Lai

Machine Learning and Knowledge Extraction, Journal Year: 2024, Volume and Issue: 6(4), P. 2321 - 2335

Published: Oct. 14, 2024

Traditional methods of agricultural disease detection rely primarily on manual observation, which is not only time-consuming and labor-intensive, but also prone to human error. The advent deep learning has revolutionized plant by providing more accurate efficient solutions. management potato diseases critical the industry, as these can lead substantial losses in crop production. prompt identification classification leaf are essential mitigating such losses. In this paper, we present a novel approach that integrates lightweight convolutional neural network architecture, RegNetY-400MF, with transfer techniques accurately identify seven different types diseases. proposed method enhances precision reduces computational storage demands, mere 0.40 GFLOPs model size 16.8 MB. This makes it well-suited for use edge devices limited resources, enabling real-time environments. experimental results demonstrated accuracy identifying was 90.68%, comprehensive solution management.

Language: Английский

Research Progress of All‐Fiber Optic Current Transformers in Novel Power Systems: A Review DOI Open Access

Zhenhua Li,

Jiuxi Cui, Haoyu Chen

et al.

Microwave and Optical Technology Letters, Journal Year: 2025, Volume and Issue: 67(1)

Published: Jan. 1, 2025

ABSTRACT The new electric power system, dominated by renewable energy sources, demands current transformers with wide bandwidth and broad dynamic sensing capabilities. An all‐fiber optic that combines fiber technology the Faraday magneto‐optical effect offers an effective solution for precise sensing. paper first introduces principle basic optical path structure of transformer (AFOCT), followed a discussion on error factors affecting measurement performance operational reliability AFOCT. It then summarizes presents specific solutions developed over past decade. Lastly, concludes summary future outlook applying AFOCT in grids. Optical are currently widely used ultrahigh extra‐high voltage transmission engineering. As matures, coupled advancements intelligence levels, prospects field promising.

Language: Английский

Citations

5
Instance Segmentation and 3D Pose Estimation of Tea Bud Leaves for Autonomous Harvesting Robots DOI Creative Commons

Haoxin Li,

Tianci Chen,

Yingmei Chen

et al.

Agriculture, Journal Year: 2025, Volume and Issue: 15(2), P. 198 - 198

Published: Jan. 17, 2025

In unstructured tea garden environments, accurate recognition and pose estimation of bud leaves are critical for autonomous harvesting robots. Due to variations in imaging distance, exhibit diverse scale characteristics camera views, which significantly complicates the process. This study proposes a method using an RGB-D precise leaves. The approach first constructs leaves, followed by dynamic weight strategy achieve adaptive estimation. Quantitative experiments demonstrate that instance segmentation model achieves mAP@50 92.0% box detection 91.9% mask detection, improving 3.2% 3.4%, respectively, compared YOLOv8s-seg model. results indicate maximum angular error 7.76°, mean 3.41°, median 3.69°, absolute deviation 1.42°. corresponding distance errors 8.60 mm, 2.83 2.57 0.81 further confirming accuracy robustness proposed method. These can be applied environments non-destructive with bud-leave

Language: Английский

Citations

1
A Lettpoint-Yolov11l Based Intelligent Robot for Precision Intra-Row Weeds Control in Lettuce DOI

Rui-Feng Wang,

Yu-Hao Tu,

Zizhong Chen

et al.

Published: Jan. 1, 2025

Language: Английский

Citations

1
Hybrid Optimization of Phase Masks: Integrating Non-Iterative Methods with Simulated Annealing and Validation via Tomographic Measurements DOI Open Access

Z. Li,

Chao Sun,

Haihua Wang

et al.

Symmetry, Journal Year: 2025, Volume and Issue: 17(4), P. 530 - 530

Published: March 31, 2025

The development of holography has facilitated significant advancements across a wide range disciplines. A phase-only spatial light modulator (SLM) plays crucial role in realizing digital holography, typically requiring phase mask as its input. Non-iterative (NI) algorithms are widely used for generation, yet they often fall short delivering precise solutions and lack adaptability complex scenarios. In contrast, the Simulated Annealing (SA) algorithm provides global optimization approach capable addressing these limitations. This study investigates integration NI with SA to enhance generation holography. Furthermore, we examine how adjusting annealing parameters, especially cooling strategy, can significantly improve system performance symmetry. Notably, observe considerable improvement efficiency when non-iterative methods employed generate initial mask. Our method achieves perfect representation symmetry desired fields. efficacy optimized masks is evaluated through optical tomographic measurements using two-dimensional mutually unbiased bases (MUBs), resulting average similarity reaching 0.99. These findings validate effectiveness our methodin optimizing underscore potential high-precision mode recognition analysis.

Language: Английский

Citations

1
Pulling Rod Structure with Flexible Adjustment Capability for Suspending the Cold Mass of a Large-Bore MCZ Magnet System: A Novel Application Investigation DOI Creative Commons
S Wang, Hui Wang, Jianhua Liu

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(5), P. 2650 - 2650

Published: March 1, 2025

Superconducting magnets’ coils need rods for suspension in vacuum dewars to minimize heat conduction. Previously, rod sets had be custom-matched specific magnet models, hindering interchangeability. However, designing or repairing magnets required new manufacturing, which was costly and time-consuming, especially with low-conductivity composite materials. In this study, a design of multi-branch support structure various adjustment functions is evaluated, applied customized MCZ superconducting tested long period. Those obtained results show that adjustable developed custom could effectively improve material strength utilization 85.14% reduce the cross-sectional area by 16.22%. Then, leakage cut significantly. The 50 K cold shield opening also reduced 5196 mm2, lowering radiation 45%. assembly time period shortened 47 min. innovation study proved novel pulling addresses issues traditional rods.

Language: Английский

Citations

0
Research progress in near-infrared spectroscopy for detecting the quality of potato crops DOI Creative Commons
Wenjing Ren, Qingqing Jiang, Wenliang Qi

et al.

Chemical and Biological Technologies in Agriculture, Journal Year: 2025, Volume and Issue: 12(1)

Published: March 5, 2025

Language: Английский

Citations

0
Deep Learning for Sustainable Agriculture: A Systematic Review on Applications in Lettuce Cultivation DOI Open Access
Yinghe Qin,

Yu-Hao Tu,

Tao Li

et al.

Sustainability, Journal Year: 2025, Volume and Issue: 17(7), P. 3190 - 3190

Published: April 3, 2025

Lettuce, a vital economic crop, benefits significantly from intelligent advancements in its production, which are crucial for sustainable agriculture. Deep learning, core technology smart agriculture, has revolutionized the lettuce industry through powerful computer vision techniques like convolutional neural networks (CNNs) and YOLO-based models. This review systematically examines deep learning applications including pest disease diagnosis, precision spraying, pesticide residue detection, crop condition monitoring, growth stage classification, yield prediction, weed management, irrigation fertilization management. Notwithstanding significant contributions, several critical challenges persist, constrained model generalizability dynamic settings, exorbitant computational requirements, paucity of meticulously annotated datasets. Addressing these is essential improving efficiency, adaptability, sustainability learning-driven solutions production. By enhancing resource reducing chemical inputs, optimizing cultivation practices, contributes to broader goal explores research progress, optimization strategies, future directions strengthen learning’s role fostering farming.

Language: Английский

Citations

0
Advances, trends, and hotspots in sport for sustainable development: A bibliometric analysis using CiteSpace and VOSviewer DOI

Defeng Dong,

Bing He, Chen Dong

et al.

Innovation and Emerging Technologies, Journal Year: 2025, Volume and Issue: 12

Published: Jan. 1, 2025

The contribution of sports in promoting sustainable development has been widely recognized by the international community. Sustainable become core concept development. In order to systematically review evolution trend, hot spots, and future research directions promotion development, this article adopts a bibliometrics method is based on Web Science database. CiteSpace (version 6.2.R3) software VOSviewer 1.6.20) were applied perform visual analysis 565 relevant articles published between 2003 2024. Those obtained results show that growing stages, an interdisciplinary system initially formed. current focuses institutions higher learning. UK University California System are most influential countries field; China largest number publications, but cooperation relatively weak; Sustainability journal with publications field. Keyword shows events, tourism, health promotion, social inclusion equality, green sports, environmental awareness, physical education, hotspots. Interdisciplinary from perspective globalization, adaptation, action sport context world turbulence climate change, long cycle longitudinal research, digital sport, evaluation progress Development Goals (SDGs) will be directions. This study provides important reference for academics practitioners industry aims enhance awareness potential promote realization SDGs.

Language: Английский

Citations

0
Potato precision planter metering system based on improved YOLOv5n-ByteTrack DOI Creative Commons
Chao Xiao,

Changlin Song,

Junmin Li

et al.

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

Published: April 28, 2025

Accurate assessment of the planting effect is crucial during potato cultivation process. Currently, manual statistical methods are inefficient and challenging to evaluate in real-time. To address this issue, study proposes a detection algorithm for machine’s seed scooping scene, based on an improved lightweight YOLO v5n model. Initially, C3-Faster module introduced, which reduces number parameters computational load while maintaining accuracy. Subsequently, re-parameterized convolution (RepConv) incorporated into feature extraction network architecture, enhancing model’s inference speed by leveraging correlation between features. Finally, further improve efficiency model mobile applications, layer-adaptive magnitude-based pruning (LAMP) technology employed eliminate redundant channels with minimal impact performance. The experimental results indicate that: 1) YOLOv5n exhibits 56.8% reduction parameters, 56.1% decrease giga floating point operations per second (GFLOPs), 51.4% size, 37.0% Embedded Device Inference Time compared Additionally, mean average precision (mAP) at [email protected] achieves up 98.0%. 2) Compared series model, close, GFLOPs, size significantly decreased. 3) Combining ByteTrack counting method, accuracy reaches 96.6%. Based these improvements, we designed planter metering system that supports real-time monitoring omission, replanting, qualified casting This provides effective support offers visual representation outcomes, demonstrating its practical value industry.

Language: Английский

Citations

0
Optimizing Internet of Things Services Placement in Fog Computing Using Hybrid Recommendation System DOI Creative Commons

Hanen Ben Rjeb,

Layth Sliman, Hela Zorgati

et al.

Future Internet, Journal Year: 2025, Volume and Issue: 17(5), P. 201 - 201

Published: April 30, 2025

Fog Computing extends Cloud computing capabilities by providing computational resources closer to end users. has gained considerable popularity in various domains such as drones, autonomous vehicles, and smart cities. In this context, the careful selection of suitable optimal assignment services these (the service placement problem (SPP)) is essential. Numerous studies have attempted tackle issue. However, best our knowledge, none previously proposed works took into consideration dynamic context awareness user preferences for IoT placement. To deal with issue, we propose a hybrid recommendation system that combines two techniques: collaborative filtering content-based recommendation. By considering preferences, needs, resource availability, provides suggestions each service. assess efficiency system, validation scenario based on Internet Drones (IoD) was simulated tested. The results show approach leads reduction waiting time substantial improvement utilization number executed services.

Language: Английский

Citations

0