Winter Wheat SPAD Prediction Based on Multiple Preprocessing, Sequential Module Fusion, and Feature Mining Methods DOI Creative Commons

櫻井 克年,

Xiangxiang Su,

Yue Hu

и другие.

Agriculture, Год журнала: 2024, Номер 14(12), С. 2258 - 2258

Опубликована: Дек. 10, 2024

Chlorophyll is a crucial indicator for monitoring crop growth and assessing nutritional status. Hyperspectral remote sensing plays an important role in precision agriculture, offering non-destructive approach to predicting leaf chlorophyll. However, canopy spectra often face background noise data redundancy challenges. To tackle these issues, this study develops integrated processing strategy incorporating multiple preprocessing techniques, sequential module fusion, feature mining methods. Initially, the original spectrum (OS) from 2021, 2022, fusion year underwent through Fast Fourier Transform (FFT) smoothing, scattering correction (MSC), first derivative (FD), second (SD). Secondly, was conducted using Competitive Adaptive Reweighted Sampling (CARS), Iterative Retention of Information Variables (IRIV), Principal Component Analysis (PCA) based on optimal order data. Finally, Partial Least Squares Regression (PLSR) used construct prediction model winter wheat SPAD compare effects different years stages. The findings show that FFT-MSC (firstly pre-processing FFT, secondly secondary FFT spectral MSC) effectively reduced issues such as noisy signals baseline drift. FFT-MSC-IRIV-PLSR (based combined preprocessed data, screening IRIV, then combining with PLSR model) predicts highest overall accuracy, R2 0.79–0.89, RMSE 4.51–5.61, MAE 4.01–4.43. performed best 0.84–0.89 4.51–6.74. during stages occurred early filling stage, 0.75 0.58. On basis research, future work will focus optimizing process richer environmental so further enhance predictive capability applicability model.

Язык: Английский

Enhancing Winter Wheat Soil–Plant Analysis Development Value Prediction through Evaluating Unmanned Aerial Vehicle Flight Altitudes, Predictor Variable Combinations, and Machine Learning Algorithms DOI Creative Commons
Jianjun Wang, Yin Quan,

Lige Cao

и другие.

Plants, Год журнала: 2024, Номер 13(14), С. 1926 - 1926

Опубликована: Июль 12, 2024

Monitoring winter wheat Soil-Plant Analysis Development (SPAD) values using Unmanned Aerial Vehicles (UAVs) is an effective and non-destructive method. However, predicting SPAD during the booting stage less accurate than other growth stages. Existing research on UAV-based value prediction has mainly focused low-altitude flights of 10-30 m, neglecting potential benefits higher-altitude flights. The study evaluates predictions Vegetation Indices (VIs) from UAV images at five different altitudes (i.e., 20, 40, 60, 80, 100, 120 respectively, a DJI P4-Multispectral as example, with resolution 1.06 to 6.35 cm/pixel). Additionally, we compare predictive performance various predictor variables (VIs, Texture (TIs), Discrete Wavelet Transform (DWT)) individually in combination. Four machine learning algorithms (Ridge, Random Forest, Support Vector Regression, Back Propagation Neural Network) are employed. results demonstrate comparable between m (with cm/pixel) 20 This finding significantly improves efficiency monitoring since flying UAVs higher greater coverage, thus reducing time needed for scouting when same heading overlap side rates. overall trend accuracy follows: VIs + TIs DWT > DWT. set obtains frequency information (DWT), compensating limitations set. enhances effectiveness agricultural practices.

Язык: Английский

Процитировано

4

Coupling Image-Fusion Techniques with Machine Learning to Enhance Dynamic Monitoring of Nitrogen Content in Winter Wheat from UAV Multi-Source DOI Creative Commons
Xinwei Li,

Xiangxiang Su,

Jun Li

и другие.

Agriculture, Год журнала: 2024, Номер 14(10), С. 1797 - 1797

Опубликована: Окт. 12, 2024

Plant nitrogen concentration (PNC) is a key indicator reflecting the growth and development status of plants. The timely accurate monitoring plant PNC great significance for refined management crop nutrition in field. rapidly developing sensor technology provides powerful means PNC. Although RGB images have rich spatial information, they lack spectral information red edge near infrared bands, which are more sensitive to vegetation. Conversely, multispectral offer superior resolution but typically lag detail compared images. Therefore, purpose this study improve accuracy efficiency by combining advantages through image-fusion technology. This was based on booting, heading, early-filling stages winter wheat, synchronously acquiring UAV MS data, using Gram–Schmidt (GS) principal component (PC) methods generate fused evaluate them with multiple image-quality indicators. Subsequently, models predicting wheat were constructed machine-selection algorithms such as RF, GPR, XGB. results show that RGB_B1 image contains richer details other bands. GS method PC method, performance fusing high-resolution band optimal. After fusion, correlation between vegetation indices (VIs) has been enhanced varying degrees different periods, significantly enhancing response ability To comprehensively assess potential estimating PNC, fully before after fusion machine learning Random Forest (RF), Gaussian Process Regression (GPR), eXtreme Gradient Boosting (XGB). model established high stability single period, varieties, treatments, making it better than image. most significant enhancements during booting stages, particularly RF algorithm, achieved an 18.8% increase R2, 26.5% RPD, 19.7% decrease RMSE. effective technical dynamic nutritional strong support precise nutrition.

Язык: Английский

Процитировано

3

UAV Remote Sensing Technology for Wheat Growth Monitoring in Precision Agriculture: Comparison of Data Quality and Growth Parameter Inversion DOI Creative Commons
Jikai Liu, Weiqiang Wang, Jun Li

и другие.

Agronomy, Год журнала: 2025, Номер 15(1), С. 159 - 159

Опубликована: Янв. 10, 2025

The quality of the image data and potential to invert crop growth parameters are essential for effectively using unmanned aerial vehicle (UAV)-based sensor systems in precision agriculture (PA). However, existing research falls short providing a comprehensive examination inversion parameters, there is still ambiguity regarding how affects potential. Therefore, this study explored application RGB multispectral (MS) images acquired from three lightweight UAV platforms realm PA: DJI Mavic 2 Pro (M2P), Phantom 4 Multispectral (P4M), 3 (M3M). reliability pixel-scale was evaluated based on assessment metrics, winter wheat above-ground biomass (AGB), plant nitrogen content (PNC) soil analysis development (SPAD), were inverted machine learning models multi-source features at plot scale. results indicated that M3M outperformed M2P, while MS marginally superior P4M. Nevertheless, these advantages did not improve accuracy Spectral (SFs) derived P4M-based demonstrated significant AGB (R2 = 0.86, rRMSE 27.47%), SFs M2P-based camera exhibited best performance SPAD 0.60, 7.67%). Additionally, combining spectral textural yielded highest PNC 0.82, 14.62%). This clarified prevalent mounted PA their influence parameter potential, offering guidance selecting appropriate sensors monitoring key parameters.

Язык: Английский

Процитировано

0

Precision Agriculture: Temporal and Spatial Modeling of Wheat Canopy Spectral Characteristics DOI Creative Commons
Donghui Zhang, Liang Hou,

Liangjie Lv

и другие.

Agriculture, Год журнала: 2025, Номер 15(3), С. 326 - 326

Опубликована: Фев. 1, 2025

This study investigates the dynamic changes in wheat canopy spectral characteristics across seven critical growth stages (Tillering, Pre-Jointing, Jointing, Post-Jointing, Booting, Flowering, and Ripening) using UAV-based multispectral remote sensing. By analyzing four key bands—green (G), red (R), red-edge (RE), near-infrared (NIR)—and their combinations, we identify features that reflect activity, health, structure. Results show green band is highly sensitive to chlorophyll activity low coverage during Tillering stage, while NIR captures structural complexity density Jointing Booting stages. The combination of G bands reveals increased concentration RE effectively detects plant senescence reduced uniformity ripening stage. Time-series analysis data improves accuracy stage identification, with offering insights into inflection points. Spatially, demonstrates potential for identifying field-level anomalies, such as water stress or disease, providing actionable targeted interventions. comprehensive spatio-temporal monitoring framework crop management offers a cost-effective, precise solution disease prediction, yield forecasting, resource optimization. paves way integrating UAV sensing precision agriculture practices, future research focusing on hyperspectral integration enhance models.

Язык: Английский

Процитировано

0

Data Integration Based on UAV Multispectra and Proximal Hyperspectra Sensing for Maize Canopy Nitrogen Estimation DOI Creative Commons

Fuhao Lu,

Sun Hai-ming,

Tao Leí

и другие.

Remote Sensing, Год журнала: 2025, Номер 17(8), С. 1411 - 1411

Опубликована: Апрель 16, 2025

Nitrogen (N) is critical for maize (Zea mays L.) growth and yield, necessitating precise estimation of canopy nitrogen concentration (CNC) to optimize fertilization strategies. Remote sensing technologies, such as proximal hyperspectral sensors unmanned aerial vehicle (UAV)-based multispectral imaging, offer promising solutions non-destructive CNC monitoring. This study evaluates the effectiveness sensor UAV-based data integration in estimating spring during key stages (from 11th leaf stage, V11, Silking R1). Field experiments were conducted collect (20 vegetation indices [MVI] 24 texture [MTI]), (24 [HVI] 20 characteristic [HCI]), alongside laboratory analysis 120 samples. The Boruta algorithm identified important features from integrated datasets, followed by correlation between these Random Forest (RF)-based modeling, with SHAP (SHapley Additive exPlanations) values interpreting feature contributions. Results demonstrated model achieved high accuracy Computational Efficiency (CE) (R2 = 0.879, RMSE 0.212, CE 2.075), outperforming HVI-HCI 0.832, 0.250, =2.080). Integrating yields a high-precision 0.903, 0.190), standalone models 2.73% 8.53%, respectively. However, decreased 1.93% 1.68%, Key included red-edge (NREI, NDRE, CI) parameters (R1m), (SR, PRI) spectral (SDy, Rg) exhibited varying directional impacts on using RF. Together, findings highlight that Boruta–RF–SHAP strategy demonstrates synergistic value integrating multi-source enhancing management cultivation.

Язык: Английский

Процитировано

0

OBM-RFEcv: An adaptive ensemble model for monitoring key growth indicators of Gerbera using multi-spectral image fusion features DOI Creative Commons
Xinrui Wang, Yi Shen, Peng Tian

и другие.

PLoS ONE, Год журнала: 2025, Номер 20(5), С. e0322851 - e0322851

Опубликована: Май 20, 2025

This study aims to address the challenge of monitoring Plant Height (PH), SPAD, Leaf Area Index (LAI), and Above-Ground Biomass (AGB) in Gerbera under greenhouse cultivation conditions. We initially gathered multi-spectral images corresponding ground truth data these parameters at various growth stages using a low-altitude UAV. From collected images, we derived five Vegetation Indices (VIs): NDVI, GNDVI, LCI, NDRE, OSAVI, extracted their textural features as fusion features. An adaptive ensemble model, OBM-RFEcv, was then developed by integrating six base models (Linear Regression, Decision Tree Regressor, Random Forest XGBoost Support Vector Regressor) with Recursive Feature Elimination (RFE) predict key indicators. The results indicate that OBM-RFEcv model outperforms other when VIs, particularly test dataset, where it achieved highest accuracy for PH (NDVI), SPAD (GNDVI), LAI AGB (NDRE) R 2 values 0.92, 0.90, 0.89, 0.93, respectively. root mean square error (RMSE) were 0.04, 0.07, 0.08, respectively, showing improvements over best individual 0.01, 0.03, 0.09 , reductions RMSE These findings confirm based on image fusion, effectively monitors indicators Gerbera, providing non-invasive precise method crop monitoring.

Язык: Английский

Процитировано

0

Precision estimation of winter wheat crop height and above-ground biomass using unmanned aerial vehicle imagery and oblique photoghraphy point cloud data DOI Creative Commons
Yafeng Li, Changchun Li, Qian Cheng

и другие.

Frontiers in Plant Science, Год журнала: 2024, Номер 15

Опубликована: Сен. 18, 2024

Introduction Crop height and above-ground biomass (AGB) serve as crucial indicators for monitoring crop growth estimating grain yield. Timely accurate acquisition of wheat AGB data is paramount guiding agricultural production. However, traditional methods suffer from drawbacks such time-consuming, laborious destructive sampling. Methods The current approach to using unmanned aerial vehicles (UAVs) remote sensing relies solely on spectral data, resulting in low accuracy estimation. This method fails address the ill-posed inverse problem mapping two-dimensional three-dimensional issues related saturation. To overcome these challenges, RGB multispectral sensors mounted UAVs were employed acquire image data. five-directional oblique photography technique was utilized construct point cloud extracting height. Results Discussion study comparatively analyzed potential mean Accumulated Incremental Height (AIH) extraction. Utilizing Vegetation Indices (VIs), AIH their feature combinations, models including Random Forest Regression (RFR), eXtreme Gradient Boosting (XGBoost), Trees (GBRT), Support Vector (SVR) Ridge (RR) constructed estimate winter AGB. research results indicated that performed well extraction, with minimal differences between 95% measured values observed across various stages wheat, yielding R 2 ranging 0.768 0.784. Compared individual features, combination multiple features significantly improved model’s accuracy. incorporation helps alleviate effects Coupling VIs increases 0.694-0.885 only 0.728-0.925. In comparing performance five machine learning algorithms, it discovered based decision trees superior other algorithms. Among them, RFR algorithm optimally, 0.9 0.93. Conclusion conclusion, leveraging multi-source algorithms overcomes limitations methods, offering a technological reference precision agriculture management decision-making.

Язык: Английский

Процитировано

1

Impact of remote sensing data fusion on agriculture applications: A review DOI
Ayyappa Reddy Allu,

Shashi Mesapam

European Journal of Agronomy, Год журнала: 2024, Номер 164, С. 127478 - 127478

Опубликована: Дек. 18, 2024

Язык: Английский

Процитировано

1

Winter Wheat SPAD Prediction Based on Multiple Preprocessing, Sequential Module Fusion, and Feature Mining Methods DOI Creative Commons

櫻井 克年,

Xiangxiang Su,

Yue Hu

и другие.

Agriculture, Год журнала: 2024, Номер 14(12), С. 2258 - 2258

Опубликована: Дек. 10, 2024

Chlorophyll is a crucial indicator for monitoring crop growth and assessing nutritional status. Hyperspectral remote sensing plays an important role in precision agriculture, offering non-destructive approach to predicting leaf chlorophyll. However, canopy spectra often face background noise data redundancy challenges. To tackle these issues, this study develops integrated processing strategy incorporating multiple preprocessing techniques, sequential module fusion, feature mining methods. Initially, the original spectrum (OS) from 2021, 2022, fusion year underwent through Fast Fourier Transform (FFT) smoothing, scattering correction (MSC), first derivative (FD), second (SD). Secondly, was conducted using Competitive Adaptive Reweighted Sampling (CARS), Iterative Retention of Information Variables (IRIV), Principal Component Analysis (PCA) based on optimal order data. Finally, Partial Least Squares Regression (PLSR) used construct prediction model winter wheat SPAD compare effects different years stages. The findings show that FFT-MSC (firstly pre-processing FFT, secondly secondary FFT spectral MSC) effectively reduced issues such as noisy signals baseline drift. FFT-MSC-IRIV-PLSR (based combined preprocessed data, screening IRIV, then combining with PLSR model) predicts highest overall accuracy, R2 0.79–0.89, RMSE 4.51–5.61, MAE 4.01–4.43. performed best 0.84–0.89 4.51–6.74. during stages occurred early filling stage, 0.75 0.58. On basis research, future work will focus optimizing process richer environmental so further enhance predictive capability applicability model.

Язык: Английский

Процитировано

0