Machine Learning and Spatio Temporal Analysis for Assessing Ecological Impacts of the Billion Tree Afforestation Project

Ecology and Evolution, Год журнала: 2025, Номер 15(2)

Опубликована: Фев. 1, 2025

ABSTRACT This study evaluates the Billion Tree Afforestation Project (BTAP) in Pakistan's Khyber Pakhtunkhwa (KPK) province using remote sensing and machine learning. Applying Random Forest (RF) classification to Sentinel‐2 imagery, we observed an increase tree cover from 25.02% 2015 29.99% 2023 a decrease barren land 20.64% 16.81%, with accuracy above 85%. Hotspot spatial clustering analyses revealed significant vegetation recovery, high‐confidence hotspots rising 36.76% 42.56%. A predictive model for Normalized Difference Vegetation Index (NDVI), supported by SHAP analysis, identified soil moisture precipitation as primary drivers of growth, ANN achieving R 2 0.8556 RMSE 0.0607 on testing dataset. These results demonstrate effectiveness integrating learning framework support data‐driven afforestation efforts inform sustainable environmental management practices.

Язык: Английский

---

Like draws to like: A Multi-granularity Ball-Intra Fusion approach for fault diagnosis models to resists misleading by noisy labels

Advanced Engineering Informatics, Год журнала: 2024, Номер 60, С. 102425 - 102425

Опубликована: Фев. 23, 2024

Язык: Английский

---

MCFNet: Research on small target detection of RGB‐infrared under UAV perspective with multi‐scale complementary feature fusion

IET Image Processing, Год журнала: 2025, Номер 19(1)

Опубликована: Янв. 1, 2025

Abstract The task of single‐modal image target detection from a drone's perspective faces multiple challenges such as large sizes, small and dense targets, insufficient lighting conditions, hardware resource constraints, all which affect the accuracy real‐time performance algorithms. In response, algorithm that fuses multi‐scale complementary features, named MCFNet is proposed. Firstly, in order to independently extract unimodal features visible light infrared images effectively enhance information between images, dual‐stream backbone network with terminal fusion mechanism Secondly, feature residual module utilized optimize integration process features. Subsequently, by designing enhancement network's capability capture enhanced. Finally, for targets varying lightweight Transformer extraction proposed improve perspective. Test results on drone‐vehicle dataset show this method achieved an average 67.92%, while self‐constructed UAV‐data reached 96.4%. Additionally, series ablation experiments validated effectiveness different modules.

Язык: Английский

---

Complex Evidence Theory for Multisource Data Fusion

Chinese journal of information fusion., Год журнала: 2024, Номер 1(2), С. 134 - 159

Опубликована: Сен. 30, 2024

Data fusion is a prevalent technique for assembling imperfect raw data coming from multiple sources to capture reliable and accurate information. Dempster–Shafer evidence theory one of useful methodologies in the uncertain multisource The existing literature lacks thorough comprehensive review recent advances Dempster– Shafer fusion. Therefore, state art has be surveyed gain insight into how beneficial it evolved over time. In this paper, we first provide methods based on its extensions, collectively referred as classical theory, three aspects uncertainty modeling, fusion, decision making. Next, study explore complex both closed world open contexts that benefits frame plane modelling. We then present framework-based algorithms, which are applied pattern classification compare demonstrate their applicabilities. research results indicate framework can enhance capabilities modeling reasoning by generating constructive interference through appropriate basic belief assignment functions modeled numbers. Through analysis comparison, finally propose several challenges identify future directions theorybased

Язык: Английский

---

VFL3D: A Single-Stage Fine-Grained Lightweight Point Cloud 3D Object Detection Algorithm Based on Voxels

IEEE Transactions on Intelligent Transportation Systems, Год журнала: 2024, Номер 25(9), С. 12034 - 12048

Опубликована: Март 29, 2024

In this work, we propose a voxel-based single-stage fine-grained and efficient point cloud 3D object detection algorithm to address the inadequate granularity in feature extraction tasks imbalance between efficiency accuracy scenarios. We develop lightweight multibranch cross-sparse convolution network (LMCCN) that is designed preserve of original while achieving enhanced efficiency. Additionally, introduce compact self-attention augmented bird's eye view (BEV) module (CFSAM). This aims further refine BEV features, enabling acquisition both locally globally features thereby augmentingthe perceptual capabilities constructed model. Without bells whistles, proposed method attains excellent performance on many autonomous driving benchmarks, with accuracies up 81.67% KITTI, 72.74% ONCE, 84.00% nuScenes. Moreover, it reaches peak speed 46.08 FPS, effectively balancing speed.

Язык: Английский

---

Application of G-Ori metamaterials as sports equipment baseball bat in an electro-magneto-elastic sandwich composite beam

Mechanics of Advanced Materials and Structures, Год журнала: 2024, Номер unknown, С. 1 - 20

Опубликована: Окт. 15, 2024

In this work, a higher order thickness stretched model is employed for electro-magneto-elastic results of shear deformable baseball bat as sandwich beam. The formulation performed using the two dimensional constitutive relations well electric potential and magnetic induction bat. proposed can be used in sport equipment. principle virtual work to derive governing equations presence thermal, electrical loads. physical beam composed graphene origami reinforced copper matric core integrated with piezoelectric/piezomagnetic face-sheets thermal environment excited by potentials.

Язык: Английский

---

Smart Microneedles in Biomedical Engineering: Harnessing Stimuli‐Responsive Polymers for Novel Applications

Polymers for Advanced Technologies, Год журнала: 2024, Номер 35(12)

Опубликована: Дек. 1, 2024

ABSTRACT This review aims to provide a comprehensive analysis of recent advancements in smart microneedles (MNs) within the biomedical field, focusing on integration stimuli‐responsive polymers for enhanced therapeutic and diagnostic applications. Conventional drug delivery methods are known face limitations precision, safety, patient compliance, which can be addressed by innovative features MNs. Through use various polymers, these MNs have been designed react environmental or physiological cues, allowing on‐demand release, biomarker sensing, localized interventions. Fundamental materials used fabrication MNs, including metals, composite hydrogels, reviewed, different categories stimuli‐responsiveness, such as photo, electro, thermal, mechanical, biochemical, explored. Application‐specific designs areas delivery, cancer therapy, diabetes management, skin disease treatments also examined. this discussion, it is highlighted that poised play significant role advancing personalized noninvasive medical treatments.

Язык: Английский

---

Quantitative analysis on the optical kerr impact and third harmonic generation in beltrami-shaped curved graphene

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Янв. 11, 2025

We present a comprehensive analysis of the optical attributes graphene sheets with charge carriers residing on curved substrate. In particular, we focus fascinating case Beltrami geometry and provide an explicit parametrization for this two-dimensional surface. By employing massless Dirac description that is characteristic graphene, investigate impact properties within sample. Our findings reveal system are highly sensitive to several key factors, namely radius, surface chemical potential relaxation time. The numerical demonstrate characteristics Beltrami-shaped differ significantly from those regular due geometry. This distinction opens up exciting possibilities exploring new phenomena designing graphene-based devices customized properties.

Язык: Английский

---

A method for enhancing the accuracy of industrial bearing surface defect detection

Nondestructive Testing And Evaluation, Год журнала: 2025, Номер unknown, С. 1 - 26

Опубликована: Янв. 12, 2025

Deep learning models often encounter challenges when detecting surface defects on industrial bearings, including inaccurate feature extraction caused by random shapes and multi-scale defects, as well high leakage rates of small objects in complex multi-object scenes. To address these challenges, a YOLOv8n-based method (YOLOv8-IWDD) is proposed for bearing defects. Firstly, the ImplicitHead detection head introduced, which utilises implicit addition multiplication to highlight key features. This enhancement improves extraction, resulting improved accuracy objects, especially scenarios, effectively reduces missed detections. Secondly, Inner-WIoU loss function optimises bounding box localisation, enabling model better handle randomly shaped accelerate convergence, enhance localisation accuracy. In addition, dynamic label assignment sampling strategies are achieve flexible adjustment sample method, further robustness. These model's robustness handling shape The experimental results show that YOLOv8-IWDD mAP0.5 2.9%, mAP0.5:0.95 9.4%, precision 2.2%, recall 6.5% compared with YOLOv8n model. achieves promising bearings.

Язык: Английский

---

AM-Net: A Network with Attention and Multi-scale Feature Fusion for Skin Lesion Segmentation

IEEE Sensors Journal, Год журнала: 2025, Номер 25(5), С. 8645 - 8655

Опубликована: Янв. 20, 2025

Язык: Английский

---