Point-to-Interval Prediction Method for Key Soil Property Contents Utilizing Multi-Source Spectral Data DOI Creative Commons
Shuyan Liu, Dongyan Huang, Lili Fu

et al.

Agronomy, Journal Year: 2024, Volume and Issue: 14(11), P. 2678 - 2678

Published: Nov. 14, 2024

Key soil properties play pivotal roles in shaping crop growth and yield outcomes. Accurate point prediction interval of serve as crucial references for making informed decisions regarding fertilizer applications. Traditional testing methods often entail laborious resource-intensive chemical analyses. To address this challenge, study introduced a novel approach leveraging spectral data fusion techniques to forecast key properties. The initial datasets were derived from UV–visible–near-infrared (UV-Vis-NIR) mid-infrared (MIR) data, which underwent preprocessing stages involving smoothing denoising fractional-order derivative[s] (FOD) transform techniques. After extracting the characteristic bands both types three strategies developed, further enhanced using machine learning Among these strategies, outer-product analysis algorithm proved particularly effective improving accuracy. For predictions, metrics such coefficient determination (R2) error demonstrated significant enhancements compared predictions based solely on single-source data. Specifically, R2 values increased by 0.06 0.41, underscoring efficacy combined with partial least squares regression (PLSR). In addition, coverage width criterion establish reliable intervals properties, including organic matter (SOM), total nitrogen (TN), hydrolyzed (HN), available potassium (AK). These developed within framework kernel density estimation (KDE) model, facilitates quantification uncertainty property estimates. phosphorus (AP), preliminary assessment its concentration was also provided. By integrating advanced learning, paves way more agricultural decision sustainable management strategies.

Language: Английский

Automatic Grape Cluster Detection Combining YOLO Model and Remote Sensing Imagery DOI Creative Commons
Ana María Codes Alcaraz, Nicola Furnitto,

Giuseppe Sottosanti

et al.

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(2), P. 243 - 243

Published: Jan. 11, 2025

Precision agriculture has recently experienced significant advancements through the use of technologies such as unmanned aerial vehicles (UAVs) and satellite imagery, enabling more efficient precise agricultural management. Yield estimation from these is essential for optimizing resource allocation, improving harvest logistics, supporting decision-making sustainable vineyard This study aimed to evaluate grape cluster numbers estimated by using YOLOv7x in combination with images obtained UAVs a vineyard. Additionally, capability several vegetation indices calculated Sentinel-2 PlanetScope satellites estimate clusters was evaluated. The results showed that application model RGB acquired able accurately predict (R2 value RMSE 0.64 0.78 vine−1). On contrary, indexes derived were found not lower than 0.23), probably due fact are related vigor, which always yield parameters (e.g., number). suggests high-resolution UAV images, multispectral advanced detection models like can significantly improve accuracy management, resulting agriculture.

Language: Английский

Citations

0
Modelling soil organic carbon at multiple depths in woody encroached grasslands using integrated remotely sensed data DOI Creative Commons

Sfundo Mthiyane,

Onisimo Mutanga, Trylee Nyasha Matongera

et al.

Environmental Monitoring and Assessment, Journal Year: 2025, Volume and Issue: 197(3)

Published: March 1, 2025

Abstract Woody plants encroachment into grasslands has considerable hydrological and biogeochemical consequences to grassland soils that include altering the Soil Organic Carbon (SOC) pool. Consequently, continuous SOC stock assessment evaluation at deeper soil depths of woody encroached is essential for informed management monitoring phenomenon. Due high litter biomass deep root structures, landscapes have been suggested alter accumulation layers; however, extent which sequester within localized protected still poorly understood. Remote sensing methods techniques recently popular in analysis due better spatial spectral data properties as well availability affordable eco-friendly data. In this regard, study sought quantify various (30 cm, 60 100 cm) a woody-encroached by integrating Sentinel-1 (S1), Sentinel-2 (S2), PlanetScope (PS) satellite imagery, topographic variables. was quantified from 360 field-collected samples using loss-On-Ignition (LOI) method distribution across Bisley Nature Reserve modelled employing Random Forest (RF) algorithm. The study’s results demonstrate integration variables, Synthetic Aperture Radar (SAR), effectively stocks all investigated depths, with R 2 values 0.79 RMSE 0.254 t/ha. Interestingly, were higher 30 cm compared depths. horizontal reception (VH), Slope, Topographic Weightiness Index (TWI), Band 11 vertical (VV) optimal predictors landscapes. These highlight significance RF model variables accurate modelling ecosystems. findings are pivotal developing cost-effective labour-efficient system appropriate habitats.

Language: Английский

Citations

0
Spatial Inversion of Soil Organic Carbon Content Based on Hyperspectral Data and Sentinel‐2 Images DOI

Xiaoyu Huang,

Xuemei Wang, Yanping Guo

et al.

Land Degradation and Development, Journal Year: 2025, Volume and Issue: unknown

Published: April 3, 2025

ABSTRACT Given that Sentinel‐2 (S2) multispectral images provide extensive spatial information and ground‐based hyperspectral data capture refined spectral characteristics, their integration can enhance both the comprehensiveness precision of surface acquisition. This study seeks to leverage these sources develop an optimized estimation model for accurately monitoring large‐scale soil organic carbon (SOC) content, thereby addressing current limitations in multi‐source fusion research. In this study, using mathematical transformation discrete wavelet transform process ground delta oasis Weigan Kuqa rivers Xinjiang, China, combination with S2 image, machine learning algorithms were employed construct models SOC content total variables characteristic variables, inversion oases was carried out. We found R ‐DWT‐H9 significantly correlation between ( p < 0.001). The accuracy constructed based on feature selected by SPA IRIV generally higher than variable models. IRIV‐RFR had highest stable capability. values 2 training validation sets 0.66 0.64, respectively. RMSE 1.5 g∙kg −1 , RPD > 1.4. interior oasis, mainly deficient (61.35%) or relatively (8.17%), while periphery it extremely (30.48%). Combine providing a reference evaluating fertility arid regions.

Language: Английский

Citations

0
Point-to-Interval Prediction Method for Key Soil Property Contents Utilizing Multi-Source Spectral Data DOI Creative Commons
Shuyan Liu, Dongyan Huang, Lili Fu

et al.

Agronomy, Journal Year: 2024, Volume and Issue: 14(11), P. 2678 - 2678

Published: Nov. 14, 2024

Key soil properties play pivotal roles in shaping crop growth and yield outcomes. Accurate point prediction interval of serve as crucial references for making informed decisions regarding fertilizer applications. Traditional testing methods often entail laborious resource-intensive chemical analyses. To address this challenge, study introduced a novel approach leveraging spectral data fusion techniques to forecast key properties. The initial datasets were derived from UV–visible–near-infrared (UV-Vis-NIR) mid-infrared (MIR) data, which underwent preprocessing stages involving smoothing denoising fractional-order derivative[s] (FOD) transform techniques. After extracting the characteristic bands both types three strategies developed, further enhanced using machine learning Among these strategies, outer-product analysis algorithm proved particularly effective improving accuracy. For predictions, metrics such coefficient determination (R2) error demonstrated significant enhancements compared predictions based solely on single-source data. Specifically, R2 values increased by 0.06 0.41, underscoring efficacy combined with partial least squares regression (PLSR). In addition, coverage width criterion establish reliable intervals properties, including organic matter (SOM), total nitrogen (TN), hydrolyzed (HN), available potassium (AK). These developed within framework kernel density estimation (KDE) model, facilitates quantification uncertainty property estimates. phosphorus (AP), preliminary assessment its concentration was also provided. By integrating advanced learning, paves way more agricultural decision sustainable management strategies.

Language: Английский

Citations

0