Advances in vegetation mapping through remote sensing and machine learning techniques: a scientometric review

European Journal of Remote Sensing, Journal Year: 2024, Volume and Issue: 57(1)

Published: Oct. 30, 2024

This study explores the rapid growth in remote-sensing technologies for vegetation mapping, driven by integration of advanced machine learning techniques. An analysis publication trends from Scopus indicates significant expansion 2019 to 2023, reflecting technological advancements and improved accessibility. Incorporating algorithms like random forest, support vector machines, neural networks, XGBRFClassifier has enhanced monitoring dynamics at various scales. progress supports addressing global environmental challenges such as climate change providing timely data conservation strategies. China leads research output, followed United States India, underscoring field's significance. Key journals, including "Remote Sensing," conferences IGARSS, play pivotal roles disseminating findings. The majority publications are articles, emphasizing reliance on original empirical data. multidisciplinary nature is evident, with contributions spanning Earth Sciences, Agriculture, Environmental Science, Computer Science. Visualisations using VOSviewer reveal interconnected themes, highlighting topics land use, change, aboveground biomass. findings emphasise importance continued international collaboration develop innovative solutions sustainability.

Language: Английский

---

Deep learning in multi-sensor agriculture and crop management

Elsevier eBooks, Journal Year: 2025, Volume and Issue: unknown, P. 335 - 379

Published: Jan. 1, 2025

Language: Английский

---

A Multi-Source Strategy for Assessing Major Winter Crops Performance and Irrigation Water Requirements

Land, Journal Year: 2025, Volume and Issue: 14(2), P. 340 - 340

Published: Feb. 7, 2025

Accurate regional crop classification, acreage estimation, yield prediction, and water requirement assessment are essential for effective agricultural planning market forecasts. This study uses an integrated geospatial statistical approach to assess major winter crops wheat sugarcane cultivation in Ghotki District, Pakistan, from 2017/18 2022/23. It combines satellite data Landsat 8 Sentinel-2, ground truthing, reporting records analyze key factors such as area, gradients, vegetation health, normalized difference index (NDVI)-based models, requirements, total irrigation consumption. Results showed that areas ranged 15% 19%, with the highest coverage observed 2021/22 season. Sugarcane 6% 10%, peaking 2018/19 A strong linear association between NDVI (R2 = 0.86) was observed. Wheat predictions utilized regression, robust regression all of which were validated by findings. Irrigation demand season calculated at 1887 million cubic meters (MCM) 2017/18, 1357 MCM supplied Sindh Drainage Authority (SIDA). By 2020/21, reached 2023 MCM, while SIDA’s supply MCM. These results highlight significance integrating analysis provide timely, reliable estimates cropped areas, forecasting, dynamics, planning. The proposed methodology contributes a scaleable solution informed decision-making resource management, applicable across other districts Pakistan on global scale.

Language: Английский

---

A Python Framework for Crop Yield Estimation Using Sentinel-2 Satellite Data

Earth, Journal Year: 2025, Volume and Issue: 6(1), P. 15 - 15

Published: March 6, 2025

Remote sensing technologies are essential for monitoring crop development and improving agricultural management. This study investigates the automation of Sentinel-2 satellite data processing to enhance wheat growth provide actionable insights smallholder farmers. The objectives include (i) analyzing vegetation indices across phenological stages refine (ii) developing a cost-effective user-friendly web application automated processing. methodology introduces “Area Under Curve” (AUC) as an independent variable yield forecasting. Among examined (NDVI, EVI, GNDVI, LAI, newly developed RE-PAP), GNDVI LAI emerged most reliable predictors yield. findings highlight importance Tillering Grain Filling stage in predictive modeling. application, integrating Python with Google Earth Engine, enables real-time monitoring, optimizing resource allocation, supporting precision agriculture. While approach demonstrates strong capabilities, further research is needed improve its generalizability. Expanding dataset diverse regions incorporating machine learning Natural Language Processing (NLP) could automation, usability, accuracy.

Language: Английский

---

Quantifying land surface changes to climatic and anthropogenic forcings by analyses of a time-series of remotely sensed images from 1936 to 2021 for a former dust bowl drought area in western Kansas, USA

Remote Sensing Applications Society and Environment, Journal Year: 2025, Volume and Issue: unknown, P. 101574 - 101574

Published: April 1, 2025

Language: Английский

---

Enhancing Land Cover Mapping in Mixed Vegetation Regions Using Remote Sensing Evapotranspiration

IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2024, Volume and Issue: 62, P. 1 - 22

Published: Jan. 1, 2024

Vegetation constitutes a significant portion of land cover. Due to the high spatial heterogeneity site conditions and similarities in spectral reflectance shapes among different vegetation types, cover mapping accuracy is often low mixed regions with multiple types. In addition traditional factors such as characteristics, topography, commonly used features, we present novel framework that incorporates evapotranspiration, which exhibits variations The proposed land-cover consists following steps: (i) estimating year-round actual evapotranspiration using remote sensing SEABL model; (ii) training classifier classify based on integrated factors, including bands, indices, night light data, evapotranspiration; (iii) generating maps account for (ETLULC); (iv) comparing evaluating variability results against input schemes existing products. typical region ETLULC demonstrates impressive performance, achieving an overall 93%. classification all types exceeds 90%. Compared methodologies do not incorporate input, significantly improves recognition cropland, forest, grassland by 5.4-15.3%, 0-15.7%, 3.0-20.4%, respectively. Moreover, strong agreement products applied Ordos Basin, particularly cropland (54.7-82.3%), forest (32.2-71.7%), (56.4-94.3%). performance underscores effectiveness this innovative framework. This study introduces approach leveraging characterized enhance mapping. method holds practical value has broad applicability identifying effective feature combinations extensively distributed regions.

Language: Английский

---

Land cover multiclass classification of wonosobo, Indonesia with time series-based one-dimensional deep learning model

Remote Sensing Applications Society and Environment, Journal Year: 2023, Volume and Issue: 32, P. 101040 - 101040

Published: Aug. 8, 2023

Language: Английский

---

Target-Aware Yield Prediction (TAYP) Model Used to Improve Agriculture Crop Productivity

IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2024, Volume and Issue: 62, P. 1 - 11

Published: Jan. 1, 2024

Because rice is the most important food crop, its yield prediction has a critical impact on policy and farmer income. In this paper, we propose new model for rice, called target-aware (TAYP) that can effectively improve accuracy of prediction. The proposed TAYP LSTM-based network, in which modify loss function by introducing target yield. Unlike traditional independent yield, our design make sensitive to such increased. To test model, use dataset from Taiwan Agricultural Research Institute, consists multispectral vegetation indexes collected drones. experimental results show performs better than related works various evaluation criteria. Compared LSTM improves RMSE R-squared 6.1% 13.0%, respectively, while increasing 89% 95%. Particularly, Kappa value 0.82 almost perfect agreement with real measurement. It clear significant improvement potential be useful tool improving agricultural productivity.

Language: Английский

---

Normalized difference vegetation index analysis reveals increase of biomass production and stability during the conversion from conventional to organic farming

Global Change Biology, Journal Year: 2024, Volume and Issue: 30(8)

Published: Aug. 1, 2024

Monitoring agriculture by remote sensing enables large-scale evaluation of biomass production across space and time. The normalized difference vegetation index (NDVI) is used as a proxy for green biomass. Here, we satellite-derived NDVI arable farms in the Netherlands to evaluate changes following conversion from conventional organic farming. We compared stability 72 fields on sand marine clay soils. Thirty-six these had been converted into between 0 50 years ago (with 2017 reference year), while other 36 were paired control where farming continued. high-resolution images Sentinel-2 satellite obtain estimates 5 (January 2016-October 2020). Overall, did not differ management during time series, but was significantly higher under management. lower sandy, clay, Organic that less than ~19 farms. However, diminished over eventually turned positive after since conversion. NDVI, averaged study, positively correlated soil Olsen-P measured samples collected 2017. conclude more stable fields, early transition can be overcome with Our study also indicates role P bioavailability plant examined benefit combining on-site measurements develop mechanistic understanding may help us navigate sustainable type agriculture.

Language: Английский

---

TCNT: a temporal convolutional network-transformer framework for advanced crop yield prediction

Journal of Applied Remote Sensing, Journal Year: 2024, Volume and Issue: 18(04)

Published: Nov. 5, 2024

Modern agricultural practices require accurate prediction of crop yields, particularly in the face a changing climate. Despite improvement estimating yield over years through machine learning (ML) algorithms, increased volatility and complexity weather patterns continue to make use conventional ML models unreliable for understanding intricate relationships. We therefore explore potential advanced ML, specifically transformer-based architecture coupled with temporal convolutional network (TCN) prediction. argue that transformers' ability model long-range dependencies within data sequences makes them well suited handle complex relationships comprehensive datasets. In addition, integration TCN would complement transformer's strengths by focusing on feature extraction. Alongside climatic variables, study integrates soil properties, moderate resolution imaging spectroradiometer (MODIS), average analyze factors influencing growth yield. The proposed TCN-transformer (TCNT) is trained evaluated using corn soybean values selected from 264 counties Illinois, Iowa, Wisconsin (the United States belt region). Furthermore, experimental results show superior performance our TCNT framework other state-of-the-art both in-season end-of-season predictions.

Language: Английский

---